open/llm-security-commons
1
0
Fork 0
Code Issues Releases 1 Activity
llm-security-commons/lexicon/injection-lexicon.json
Kjell Tore Guttormsen 344bc52adf fix(lexicon): retract two false provenance claims, reach 83/83 to source 
The lexicon was seeded from a transcribed dump. With the source module now
readable read-only at b0de0ca on the public remote, it was imported in Node
and compared entry by entry - RegExp source, flags, label, family, array
order. That check contradicted the file twice, in mirror-image ways.

The file claimed it had rewritten raw code points into \uXXXX escapes in
homoglyph:cyrillic-latin-mix and unicode:zero-width-in-word, and carried
byte_identical_to_source: false for both. The module already writes them
escaped. Nothing was ever rewritten. The stored pattern text was correct
throughout - only the account of its origin was wrong, because the dump had
rendered the module's escapes as the characters they denote and this
repository escaped them back.

The same inversion ran the other way in multi-lang:french, which stored a raw
accented Latin e where the module writes the é escape. That was the one
pattern of 83 not byte-identical to source, and it is corrected here, making
it 83/83. Both spellings are the same regular expression: compiled in Node
bare and under u, and in Python re, over accented, unaccented, uppercase and
non-matching French input, with identical match offsets. No behaviour moved.

Exactly one pattern entry differs from HEAD, and no pattern in the file
contains a non-ASCII byte any more - which matches the module, whose regex
literals are ASCII throughout. A third claim fell with them: the translation
note asserting that the leetspeak and multi-language patterns embed accented
Latin directly was wrong in both halves.

normalisations is now [] with a normalisations_note, the convention already
in signatures/secret-egress.json. A new source_fidelity block carries the
counts, the method, the verified class membership, and both retractions in
full - retracted, not deleted. The equivalence evidence behind the withdrawn
claim (692 Node comparisons, 236 Python) is still a true statement about raw
versus escaped; it is just no longer load-bearing.

Provenance upgraded accordingly: source_commit pinned, and the caveat saying
these checks only proved agreement with a dump is gone, because they no
longer do.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01SNMcqrfNyoLRQ7qXUFZnb9
2026-08-09 22:39:32 +02:00

973 lines
46 KiB
JSON
Raw Blame History

{
"version": "0.4.0",
"id": "injection-lexicon",
"description": "Prompt-injection pattern lexicon: the four pattern families a detector matches against untrusted text, and the severity family each belongs to. Data only. The variant building that feeds these patterns (normalisation, homoglyph folding, rot13, unicode-tag escalation) and the buried-payload heuristic are engine behaviour and are deliberately NOT described here.",
"owasp": "LLM01",
"label_format": "{class}: {detail}",
"pattern_id_space": {
"$comment": "Stable, commons-owned identifier for each pattern. It exists because an expected.json in conformance/ has to name a finding, and the two runtimes that seed this repository do not name the same pattern the same way: llm-security calls it `override: ignore previous instructions`, the Python guard calls it `override:ignore-previous`. Without a shared key no fixture can be written at all, including for the patterns whose regex is byte-identical in both.",
"field": "families[].patterns[].id",
"adopted_from": "llm-ingestion-pipeline-security src/llm_ingestion_guard/injection_lexicon.json (lexicon version 1.0, repo v0.3.4, commit 0bf0729) — the `id` field of its port of the same source table. Adopted verbatim rather than invented: the guard's port already carries both names (`id` and `desc`), so the mapping is read from verified source data rather than constructed here.",
"matching_method": "Each commons pattern was matched to a guard pattern by comparing this file's `label` to the guard's `desc`, with em-dash normalised to hyphen (the two differ only by that character, in the eight hitl-trap entries). 83/83 matched, one-to-one.",
"stability": "An id is a stable identifier. Changing one is a BREAKING change for any consumer matching on it, and requires a major bump of this file.",
"aliases": "Each pattern carries an `aliases` object naming what each seeding runtime calls it today. A runtime renames its own label by changing its alias here; the `id` does not move. The two aliases do NOT have the same evidentiary status — see alias_evidence.",
"not_yet_confirmed": "Neither llm-security nor the guard has yet agreed to this id space — both were asked via coord on 2026-08-09. Until they answer, `id` is this repository's proposal, not a ratified cross-runtime contract.",
"alias_evidence": {
"llm_ingestion_guard": {
"verified": true,
"method": "The guard's coverage matrix asserts on this exact string: coverage.py builds one case per lexicon pattern passing `pattern.id` as the expected value, and its probe compares that against the set of `label` fields on the findings the guard actually returns. The id is therefore demonstrably what a guard finding carries."
},
"llm_security": {
"verified": false,
"method": "Not established. This value is the `label` field of the pattern table, which is what the source module names the pattern — not necessarily what a llm-security FINDING carries. The finding producer was never supplied (see finding.schema.json), and the known Node finding shape uses `title`, not `label`, with the SARIF profile deriving rule ids by slugging that title. Treat this alias as the pattern's name in the source table, and confirm the finding-level string before matching on it."
}
}
},
"$comment": "Extracted without behaviour change from llm-security/scanners/lib/injection-patterns.mjs. Seeded 2026-08-09 from an operator dump delivered through the local coord mailbox; re-verified the same day against the module file itself at commit b0de0ca, which corrected two transcription artefacts the dump had introduced (see source_fidelity.retracted). Only the four pattern arrays are data. checkCognitiveLoadTrap (the buried-after-2000-chars logic) and scanForInjection's variant building are ENGINE logic that never moves here; they are absent by intent, not by omission. Array order is preserved exactly as in the module.",
"provenance": {
"source_repo": "llm-security",
"source_files": [
"scanners/lib/injection-patterns.mjs"
],
"source_exports": [
"CRITICAL_PATTERNS",
"HIGH_PATTERNS",
"MEDIUM_PATTERNS",
"HYBRID_PATTERNS"
],
"source_delivery": "operator dump 2/2, coord message from llm-security, 2026-08-09; superseded by a direct read of the module at the pinned commit below",
"source_commit": "b0de0ca6d86ce697f39669d177c2c2654c280128",
"source_remote": "ssh://git@git.fromaitochitta.com/open/llm-security.git",
"verified": "differentially, against the module file at b0de0ca - all 83 patterns compared on source, flags and label, in family and array order, 2026-08-09. See source_fidelity.",
"evidence_limits": [
"The severity the engine assigns to the hybrid family was not supplied. See families[].severity_note.",
"Whether a consumer applies these patterns to raw or to normalised text is engine behaviour and is not fixed by this file. Two runtimes matching identical patterns against differently normalised text will still disagree on identical input."
]
},
"dialect": {
"name": "ecmascript",
"$comment": "Patterns are ECMAScript regular-expression source text, exactly as the source literals spell it. Flags are declared per pattern in a `flags` field and never inlined as `(?i)` or `(?m)`; a pattern with no `flags` key carries no flags. All 83 patterns compile in Node with their declared flags, in Node with `u` added, and in Python `re` with the equivalent re.I / re.M.",
"flags": {
"i": "case-insensitive",
"m": "multiline: ^ and $ match at line boundaries"
},
"features_used": [
"negative lookahead: (?!...)",
"lookahead assertion: (?=...)",
"non-capturing groups: (?:...)",
"bounded quantifiers: {n,m}",
"unicode escapes: \\uXXXX"
],
"translation_notes": [
"Python (`re`): compile with re.I where flags contain `i` and re.M where they contain `m`. No rewriting is needed; verified by compiling all 83.",
"Nine patterns contain `\\/` - a redundant escape that a JavaScript regex LITERAL requires and that `RegExp.prototype.source` preserves. It is kept rather than normalised away because both seed runtimes accept it (Node bare, Node under `u`, and Python `re`), so removing it would break byte-identity to buy nothing. Engines that reject unknown escapes (Go `regexp`, RE2) MUST report these patterns as unsupported rather than skip them silently - a skipped pattern is an invisible false negative.",
"`\\w` and `\\b` are ASCII-only in ECMAScript without `u` but Unicode-aware in Python on `str`. Input that mixes non-ASCII word characters with these patterns can therefore match differently between runtimes. This is a real portability seam. It is recorded rather than normalised, because normalising it would change behaviour in the seed runtime.",
"No pattern in this file contains a non-ASCII byte. The source module's regex literals are pure ASCII throughout - invisible code points, Cyrillic look-alikes and accented Latin are all written as \\uXXXX in the literal, and this file reproduces that spelling. An earlier version of this note claimed the leetspeak and multi-language patterns embed accented Latin directly; that was wrong in both halves (the leetspeak patterns are ASCII-only, and the one accented pattern is escaped at source). See source_fidelity.retracted."
]
},
"normalisations": [],
"normalisations_note": "Empty by result, not by omission: all 83 patterns are byte-identical to the source module, so there is no transformation to declare. Through version 0.3.0 this array held one entry; it was retracted as false in 0.4.0. See source_fidelity.",
"source_fidelity": {
"$comment": "This file applies NO transformation to the pattern text. All 83 patterns are byte-identical to the RegExp literal source in llm-security/scanners/lib/injection-patterns.mjs at commit b0de0ca. The `normalisations` array is empty because there is nothing to declare, not because the question was skipped.",
"patterns_total": 83,
"patterns_byte_identical_to_source": 83,
"method": "The module was imported in Node from a read-only clone of the public remote at b0de0ca, and every entry compared in array order on three axes: RegExp.prototype.source, RegExp.prototype.flags, and label. Family membership, array order and per-family counts were compared at the same time. Nothing was read by eye.",
"class_membership_verified": {
"zero_width_class": [
"U+200B",
"U+200C",
"U+200D",
"U+FEFF"
],
"cyrillic_class": [
"U+0430",
"U+0435",
"U+043E",
"U+0440",
"U+0441",
"U+0456",
"U+0443"
],
"$comment": "The classes inside `unicode:zero-width-in-word` and `homoglyph:cyrillic-latin-mix`, enumerated mechanically from the module bytes. Note that the zero-width class does NOT include U+00AD even though codepoints/carriers.json lists U+00AD as a zero-width carrier, and that the Cyrillic class is a 7-member set distinct from the 13-member CYRILLIC_CONFUSABLES list in that same file. The three sets are deliberately different in the seed implementation and are NOT reconciled here."
},
"retracted": [
{
"id": "unicode-escape-invisible-and-confusable",
"was": "A `normalisations` entry, present through version 0.3.0, asserting that this file rewrote raw code points as \\uXXXX escapes in `homoglyph:cyrillic-latin-mix` and `unicode:zero-width-in-word`, and carrying byte_identical_to_source: false.",
"why_retracted": "False against the module. The source literals already write those code points as \\uXXXX; nothing was ever rewritten, and both patterns are byte-identical to source. The stored pattern text was correct all along - only the account of where it came from was wrong.",
"cause": "The seed dump rendered the module's escape sequences as the literal code points they denote. Reading that as the source spelling, this repository re-escaped them and arrived at the correct bytes by way of an incorrect story. This is transcription drift, and it is why the file is now verified against the module rather than against a dump.",
"evidence_status": "The differential result recorded for the retracted claim - Node over 173 adversarial inputs in bare and `u` mode (692 comparisons) and Python `re` over 118 inputs (236 comparisons), zero differences - stands as a true statement about raw-versus-escaped equivalence. It is simply no longer load-bearing here, because no such transformation is applied."
},
{
"id": "french-escape-inverted",
"was": "`multi-lang:french` carried the character class spelled with raw accented Latin: pr[eé]c[eé]dentes?.",
"why_retracted": "The same transcription inversion in the opposite direction. The module writes pr[e\\u00e9]c[e\\u00e9]dentes? with escapes; this was the only one of the 83 patterns not byte-identical to source. Corrected in 0.4.0 to the module's spelling, which makes the count 83/83.",
"behaviour_impact": "None. The two spellings are the same regular expression. Verified by compiling both forms and comparing match presence and offsets in Node (bare and `u` mode) and in Python `re` over accented, unaccented, uppercase and non-matching French inputs - identical results throughout."
}
]
},
"families": [
{
"id": "critical",
"source_export": "CRITICAL_PATTERNS",
"severity": "critical",
"count": 21,
"patterns": [
{
"id": "override:ignore-previous",
"label": "override: ignore previous instructions",
"pattern": "ignore\\s+(?:all\\s+)?previous",
"flags": "i",
"aliases": {
"llm_security": "override: ignore previous instructions",
"llm_ingestion_guard": "override:ignore-previous"
}
},
{
"id": "override:forget-instructions",
"label": "override: forget instructions",
"pattern": "forget\\s+your\\s+(?:instructions|rules|guidelines|directives)",
"flags": "i",
"aliases": {
"llm_security": "override: forget instructions",
"llm_ingestion_guard": "override:forget-instructions"
}
},
{
"id": "override:disregard",
"label": "override: disregard instructions",
"pattern": "disregard\\s+your",
"flags": "i",
"aliases": {
"llm_security": "override: disregard instructions",
"llm_ingestion_guard": "override:disregard"
}
},
{
"id": "override:override-instructions",
"label": "override: override instructions",
"pattern": "override\\s+your",
"flags": "i",
"aliases": {
"llm_security": "override: override instructions",
"llm_ingestion_guard": "override:override-instructions"
}
},
{
"id": "override:suspend-constraints",
"label": "override: suspend constraints",
"pattern": "suspend\\s+all\\s+constraints",
"flags": "i",
"aliases": {
"llm_security": "override: suspend constraints",
"llm_ingestion_guard": "override:suspend-constraints"
}
},
{
"id": "override:prior-constraints",
"label": "override: prior constraints",
"pattern": "all\\s+prior\\s+constraints",
"flags": "i",
"aliases": {
"llm_security": "override: prior constraints",
"llm_ingestion_guard": "override:prior-constraints"
}
},
{
"id": "spoofed-header:hash-system",
"label": "spoofed header: # SYSTEM:",
"pattern": "^#+\\s*SYSTEM\\s*:",
"flags": "m",
"aliases": {
"llm_security": "spoofed header: # SYSTEM:",
"llm_ingestion_guard": "spoofed-header:hash-system"
}
},
{
"id": "spoofed-header:inst",
"label": "spoofed header: [INST]",
"pattern": "^\\[INST\\]",
"flags": "m",
"aliases": {
"llm_security": "spoofed header: [INST]",
"llm_ingestion_guard": "spoofed-header:inst"
}
},
{
"id": "spoofed-header:pipe-system",
"label": "spoofed header: <|system|>",
"pattern": "<\\|system\\|>",
"aliases": {
"llm_security": "spoofed header: <|system|>",
"llm_ingestion_guard": "spoofed-header:pipe-system"
}
},
{
"id": "spoofed-tag:system-open",
"label": "spoofed tag: <system>",
"pattern": "<system>",
"aliases": {
"llm_security": "spoofed tag: <system>",
"llm_ingestion_guard": "spoofed-tag:system-open"
}
},
{
"id": "spoofed-tag:system-close",
"label": "spoofed tag: </system>",
"pattern": "<\\/system>",
"aliases": {
"llm_security": "spoofed tag: </system>",
"llm_ingestion_guard": "spoofed-tag:system-close"
}
},
{
"id": "spoofed-header:instructions",
"label": "spoofed header: ### Instructions:",
"pattern": "^###\\s*Instructions\\s*:",
"flags": "m",
"aliases": {
"llm_security": "spoofed header: ### Instructions:",
"llm_ingestion_guard": "spoofed-header:instructions"
}
},
{
"id": "identity:no-longer",
"label": "identity: you are no longer",
"pattern": "you\\s+are\\s+no\\s+longer",
"flags": "i",
"aliases": {
"llm_security": "identity: you are no longer",
"llm_ingestion_guard": "identity:no-longer"
}
},
{
"id": "identity:you-are-now",
"label": "identity: you are now [name]",
"pattern": "you\\s+are\\s+now\\s+(?!able|going|ready|done|running|using|looking|in\\s|at\\s|on\\s)",
"flags": "i",
"aliases": {
"llm_security": "identity: you are now [name]",
"llm_ingestion_guard": "identity:you-are-now"
}
},
{
"id": "identity:new-role",
"label": "identity: your new role",
"pattern": "your\\s+new\\s+role\\s+is",
"flags": "i",
"aliases": {
"llm_security": "identity: your new role",
"llm_ingestion_guard": "identity:new-role"
}
},
{
"id": "identity:new-directive",
"label": "identity: new primary directive",
"pattern": "your\\s+new\\s+primary\\s+directive",
"flags": "i",
"aliases": {
"llm_security": "identity: new primary directive",
"llm_ingestion_guard": "identity:new-directive"
}
},
{
"id": "identity:pretend",
"label": "identity: pretend you are",
"pattern": "pretend\\s+you\\s+are",
"flags": "i",
"aliases": {
"llm_security": "identity: pretend you are",
"llm_ingestion_guard": "identity:pretend"
}
},
{
"id": "config:ignore-claudemd",
"label": "config: ignore CLAUDE.md",
"pattern": "ignore\\s+your\\s+CLAUDE\\.md",
"flags": "i",
"aliases": {
"llm_security": "config: ignore CLAUDE.md",
"llm_ingestion_guard": "config:ignore-claudemd"
}
},
{
"id": "config:override-system-prompt",
"label": "config: override system prompt",
"pattern": "override\\s+your\\s+system\\s+prompt",
"flags": "i",
"aliases": {
"llm_security": "config: override system prompt",
"llm_ingestion_guard": "config:override-system-prompt"
}
},
{
"id": "config:bypass-safety",
"label": "config: bypass safety",
"pattern": "bypass\\s+your\\s+safety",
"flags": "i",
"aliases": {
"llm_security": "config: bypass safety",
"llm_ingestion_guard": "config:bypass-safety"
}
},
{
"id": "config:disable-filtering",
"label": "config: disable output filtering",
"pattern": "disable\\s+output\\s+filtering",
"flags": "i",
"aliases": {
"llm_security": "config: disable output filtering",
"llm_ingestion_guard": "config:disable-filtering"
}
}
]
},
{
"id": "high",
"source_export": "HIGH_PATTERNS",
"severity": "high",
"count": 32,
"patterns": [
{
"id": "normalization:prioritize-user",
"label": "normalization: prioritize user over rules",
"pattern": "always\\s+prioritize\\s+user\\s+requests?\\s+over",
"flags": "i",
"aliases": {
"llm_security": "normalization: prioritize user over rules",
"llm_ingestion_guard": "normalization:prioritize-user"
}
},
{
"id": "normalization:user-precedence",
"label": "normalization: user intent precedence",
"pattern": "user\\s+intent\\s+takes?\\s+precedence\\s+over",
"flags": "i",
"aliases": {
"llm_security": "normalization: user intent precedence",
"llm_ingestion_guard": "normalization:user-precedence"
}
},
{
"id": "normalization:claudemd-conflict",
"label": "normalization: CLAUDE.md conflict override",
"pattern": "if\\s+CLAUDE\\.md\\s+conflicts",
"flags": "i",
"aliases": {
"llm_security": "normalization: CLAUDE.md conflict override",
"llm_ingestion_guard": "normalization:claudemd-conflict"
}
},
{
"id": "normalization:skip-confirmation",
"label": "normalization: skip confirmation",
"pattern": "skip\\s+confirmation\\s+for",
"flags": "i",
"aliases": {
"llm_security": "normalization: skip confirmation",
"llm_ingestion_guard": "normalization:skip-confirmation"
}
},
{
"id": "reframing:actual-task",
"label": "reframing: your actual task",
"pattern": "your\\s+actual\\s+task\\s+is",
"flags": "i",
"aliases": {
"llm_security": "reframing: your actual task",
"llm_ingestion_guard": "reframing:actual-task"
}
},
{
"id": "reframing:real-purpose",
"label": "reframing: your real purpose",
"pattern": "your\\s+real\\s+purpose\\s+is",
"flags": "i",
"aliases": {
"llm_security": "reframing: your real purpose",
"llm_ingestion_guard": "reframing:real-purpose"
}
},
{
"id": "reframing:should-actually-do",
"label": "reframing: what you should actually do",
"pattern": "what\\s+you\\s+should\\s+actually\\s+do",
"flags": "i",
"aliases": {
"llm_security": "reframing: what you should actually do",
"llm_ingestion_guard": "reframing:should-actually-do"
}
},
{
"id": "hidden-comment:agent-directed",
"label": "hidden comment: agent-directed HTML comment",
"pattern": "<!--\\s*(?:AGENT|AI|HIDDEN|ACTUAL\\s+TASK|REAL\\s+INSTRUCTION)\\s*:",
"flags": "i",
"aliases": {
"llm_security": "hidden comment: agent-directed HTML comment",
"llm_ingestion_guard": "hidden-comment:agent-directed"
}
},
{
"id": "html-obfuscation:display-none",
"label": "html-obfuscation: display:none element with content",
"pattern": "<[^>]{1,256}style\\s*=\\s*\"[^\"]{0,256}display\\s*:\\s*none[^\"]{0,256}\"[^>]{0,256}>",
"flags": "i",
"aliases": {
"llm_security": "html-obfuscation: display:none element with content",
"llm_ingestion_guard": "html-obfuscation:display-none"
}
},
{
"id": "html-obfuscation:visibility-hidden",
"label": "html-obfuscation: visibility:hidden element",
"pattern": "<[^>]{1,256}style\\s*=\\s*\"[^\"]{0,256}visibility\\s*:\\s*hidden[^\"]{0,256}\"[^>]{0,256}>",
"flags": "i",
"aliases": {
"llm_security": "html-obfuscation: visibility:hidden element",
"llm_ingestion_guard": "html-obfuscation:visibility-hidden"
}
},
{
"id": "html-obfuscation:offscreen",
"label": "html-obfuscation: off-screen positioned element",
"pattern": "<[^>]{1,256}style\\s*=\\s*\"[^\"]{0,256}position\\s*:\\s*absolute[^\"]{0,256}-\\d{3,}px[^\"]{0,256}\"[^>]{0,256}>",
"flags": "i",
"aliases": {
"llm_security": "html-obfuscation: off-screen positioned element",
"llm_ingestion_guard": "html-obfuscation:offscreen"
}
},
{
"id": "html-obfuscation:zero-font",
"label": "html-obfuscation: zero font-size element",
"pattern": "<[^>]{1,256}style\\s*=\\s*\"[^\"]{0,256}font-size\\s*:\\s*0[^\"]{0,256}\"[^>]{0,256}>",
"flags": "i",
"aliases": {
"llm_security": "html-obfuscation: zero font-size element",
"llm_ingestion_guard": "html-obfuscation:zero-font"
}
},
{
"id": "html-obfuscation:zero-opacity",
"label": "html-obfuscation: zero opacity element",
"pattern": "<[^>]{1,256}style\\s*=\\s*\"[^\"]{0,256}opacity\\s*:\\s*0[^\"]{0,256}\"[^>]{0,256}>",
"flags": "i",
"aliases": {
"llm_security": "html-obfuscation: zero opacity element",
"llm_ingestion_guard": "html-obfuscation:zero-opacity"
}
},
{
"id": "html-obfuscation:zero-size-overflow",
"label": "html-obfuscation: zero-size overflow-hidden element",
"pattern": "<[^>]{1,256}style\\s*=\\s*\"[^\"]{0,256}(?:height|width)\\s*:\\s*0[^\"]{0,256}overflow\\s*:\\s*hidden[^\"]{0,256}\"[^>]{0,256}>",
"flags": "i",
"aliases": {
"llm_security": "html-obfuscation: zero-size overflow-hidden element",
"llm_ingestion_guard": "html-obfuscation:zero-size-overflow"
}
},
{
"id": "html-obfuscation:aria-label",
"label": "html-obfuscation: injection in aria-label attribute",
"pattern": "aria-label\\s*=\\s*\"[^\"]{0,256}(?:ignore|override|system|instruction|execute|exfiltrate)[^\"]{0,256}\"",
"flags": "i",
"aliases": {
"llm_security": "html-obfuscation: injection in aria-label attribute",
"llm_ingestion_guard": "html-obfuscation:aria-label"
}
},
{
"id": "evasion:educational",
"label": "evasion: educational purpose framing",
"pattern": "for\\s+educational\\s+purposes?\\s+only",
"flags": "i",
"aliases": {
"llm_security": "evasion: educational purpose framing",
"llm_ingestion_guard": "evasion:educational"
}
},
{
"id": "evasion:hypothetical",
"label": "evasion: hypothetical scenario framing",
"pattern": "(?:this\\s+is\\s+)?a\\s+hypothetical\\s+scenario",
"flags": "i",
"aliases": {
"llm_security": "evasion: hypothetical scenario framing",
"llm_ingestion_guard": "evasion:hypothetical"
}
},
{
"id": "evasion:fictional",
"label": "evasion: fictional setting framing",
"pattern": "in\\s+a\\s+fictional\\s+(?:setting|scenario|world)\\s+where",
"flags": "i",
"aliases": {
"llm_security": "evasion: fictional setting framing",
"llm_ingestion_guard": "evasion:fictional"
}
},
{
"id": "evasion:red-team",
"label": "evasion: red-team exercise framing",
"pattern": "(?:this\\s+is\\s+)?a\\s+red[- ]?team(?:ing)?\\s+(?:exercise|simulation|test)",
"flags": "i",
"aliases": {
"llm_security": "evasion: red-team exercise framing",
"llm_ingestion_guard": "evasion:red-team"
}
},
{
"id": "evasion:security-audit-sim",
"label": "evasion: security audit simulation framing",
"pattern": "security\\s+audit\\s+simulation",
"flags": "i",
"aliases": {
"llm_security": "evasion: security audit simulation framing",
"llm_ingestion_guard": "evasion:security-audit-sim"
}
},
{
"id": "evasion:test-pretense",
"label": "evasion: test/lab/sandbox pretense",
"pattern": "pretend\\s+(?:this\\s+is|we\\s+are\\s+in)\\s+a\\s+(?:test|lab|sandbox|simulation)",
"flags": "i",
"aliases": {
"llm_security": "evasion: test/lab/sandbox pretense",
"llm_ingestion_guard": "evasion:test-pretense"
}
},
{
"id": "evasion:thought-experiment",
"label": "evasion: thought experiment framing",
"pattern": "as\\s+a\\s+thought\\s+experiment",
"flags": "i",
"aliases": {
"llm_security": "evasion: thought experiment framing",
"llm_ingestion_guard": "evasion:thought-experiment"
}
},
{
"id": "evasion:research-purpose",
"label": "evasion: research purpose framing",
"pattern": "for\\s+(?:research|academic|scientific)\\s+purposes",
"flags": "i",
"aliases": {
"llm_security": "evasion: research purpose framing",
"llm_ingestion_guard": "evasion:research-purpose"
}
},
{
"id": "evasion:academic-context",
"label": "evasion: academic context framing",
"pattern": "in\\s+an?\\s+(?:academic|research|educational)\\s+context",
"flags": "i",
"aliases": {
"llm_security": "evasion: academic context framing",
"llm_ingestion_guard": "evasion:academic-context"
}
},
{
"id": "hitl-trap:approve-immediately",
"label": "hitl-trap: approval urgency — approve immediately",
"pattern": "(?:approve|accept|confirm)\\s+(?:this\\s+)?immediately",
"flags": "i",
"aliases": {
"llm_security": "hitl-trap: approval urgency — approve immediately",
"llm_ingestion_guard": "hitl-trap:approve-immediately"
}
},
{
"id": "hitl-trap:no-need-review",
"label": "hitl-trap: approval urgency — no need to review",
"pattern": "no\\s+need\\s+to\\s+review",
"flags": "i",
"aliases": {
"llm_security": "hitl-trap: approval urgency — no need to review",
"llm_ingestion_guard": "hitl-trap:no-need-review"
}
},
{
"id": "hitl-trap:just-approve",
"label": "hitl-trap: approval urgency — just approve",
"pattern": "(?:just|simply)\\s+(?:approve|accept|confirm)\\s+(?:it|this|everything)",
"flags": "i",
"aliases": {
"llm_security": "hitl-trap: approval urgency — just approve",
"llm_ingestion_guard": "hitl-trap:just-approve"
}
},
{
"id": "hitl-trap:suppress-review",
"label": "hitl-trap: approval urgency — suppress review",
"pattern": "(?:don'?t|do\\s+not)\\s+(?:bother\\s+)?(?:check|review|verify|inspect|read)(?:ing)?\\s+(?:this|the|it)",
"flags": "i",
"aliases": {
"llm_security": "hitl-trap: approval urgency — suppress review",
"llm_ingestion_guard": "hitl-trap:suppress-review"
}
},
{
"id": "hitl-trap:brevity-skip",
"label": "hitl-trap: summary suppression — for brevity, skipping",
"pattern": "for\\s+brevity[\\s,]+(?:skipping|omitting|leaving\\s+out)",
"flags": "i",
"aliases": {
"llm_security": "hitl-trap: summary suppression — for brevity, skipping",
"llm_ingestion_guard": "hitl-trap:brevity-skip"
}
},
{
"id": "hitl-trap:too-long",
"label": "hitl-trap: summary suppression — too long to show",
"pattern": "(?:too\\s+(?:long|verbose|detailed)\\s+to\\s+(?:show|display|list|include))",
"flags": "i",
"aliases": {
"llm_security": "hitl-trap: summary suppression — too long to show",
"llm_ingestion_guard": "hitl-trap:too-long"
}
},
{
"id": "hitl-trap:details-omitted",
"label": "hitl-trap: summary suppression — details omitted",
"pattern": "(?:details?\\s+(?:omitted|redacted|hidden|suppressed))",
"flags": "i",
"aliases": {
"llm_security": "hitl-trap: summary suppression — details omitted",
"llm_ingestion_guard": "hitl-trap:details-omitted"
}
},
{
"id": "hitl-trap:scope-minimization",
"label": "hitl-trap: scope minimization — minor change + destructive action",
"pattern": "(?:minor|trivial|small|cosmetic)\\s+(?:change|update|fix|adjustment)\\b[^.]{0,80}(?:delet|remov|drop|truncat|reset|destroy|\\bformat\\b|wipe)",
"flags": "i",
"aliases": {
"llm_security": "hitl-trap: scope minimization — minor change + destructive action",
"llm_ingestion_guard": "hitl-trap:scope-minimization"
}
}
]
},
{
"id": "medium",
"source_export": "MEDIUM_PATTERNS",
"severity": "medium",
"count": 22,
"patterns": [
{
"id": "leetspeak:ignore-previous",
"label": "leetspeak: ignore previous (obfuscated)",
"pattern": "[1!|l][gq9]n[o0]r[e3]\\s+(?:all\\s+)?pr[e3]v[i1!|l][o0]us",
"flags": "i",
"aliases": {
"llm_security": "leetspeak: ignore previous (obfuscated)",
"llm_ingestion_guard": "leetspeak:ignore-previous"
}
},
{
"id": "leetspeak:forget-instructions",
"label": "leetspeak: forget instructions (obfuscated)",
"pattern": "f[o0]rg[e3]t\\s+y[o0]ur\\s+[i1!|l]nstruct[i1!|l][o0]ns",
"flags": "i",
"aliases": {
"llm_security": "leetspeak: forget instructions (obfuscated)",
"llm_ingestion_guard": "leetspeak:forget-instructions"
}
},
{
"id": "leetspeak:disregard",
"label": "leetspeak: disregard your (obfuscated)",
"pattern": "d[i1!|l]sr[e3]g[a4@]rd\\s+y[o0]ur",
"flags": "i",
"aliases": {
"llm_security": "leetspeak: disregard your (obfuscated)",
"llm_ingestion_guard": "leetspeak:disregard"
}
},
{
"id": "leetspeak:override",
"label": "leetspeak: override your (obfuscated)",
"pattern": "[o0]v[e3]rr[i1!|l]d[e3]\\s+y[o0]ur",
"flags": "i",
"aliases": {
"llm_security": "leetspeak: override your (obfuscated)",
"llm_ingestion_guard": "leetspeak:override"
}
},
{
"id": "homoglyph:cyrillic-latin-mix",
"label": "homoglyph: Cyrillic-Latin mixing in adjacent characters",
"pattern": "[a-zA-Z][\\u0430\\u0435\\u043E\\u0440\\u0441\\u0456\\u0443]|[\\u0430\\u0435\\u043E\\u0440\\u0441\\u0456\\u0443][a-zA-Z]",
"aliases": {
"llm_security": "homoglyph: Cyrillic-Latin mixing in adjacent characters",
"llm_ingestion_guard": "homoglyph:cyrillic-latin-mix"
}
},
{
"id": "unicode:zero-width-in-word",
"label": "unicode: zero-width character inside word (keyword splitting)",
"pattern": "\\w[\\u200B\\u200C\\u200D\\uFEFF]\\w",
"aliases": {
"llm_security": "unicode: zero-width character inside word (keyword splitting)",
"llm_ingestion_guard": "unicode:zero-width-in-word"
}
},
{
"id": "indirect:instruction-to-ai",
"label": "indirect: instruction addressed to AI/assistant",
"pattern": "(?:note|message|instruction)\\s+(?:to|for)\\s+(?:the\\s+)?(?:AI|assistant|model|LLM|Claude)\\b",
"flags": "i",
"aliases": {
"llm_security": "indirect: instruction addressed to AI/assistant",
"llm_ingestion_guard": "indirect:instruction-to-ai"
}
},
{
"id": "indirect:direct-address",
"label": "indirect: direct address to AI/assistant",
"pattern": "(?:dear|attention)\\s+(?:AI|assistant|model|LLM|Claude)\\b",
"flags": "i",
"aliases": {
"llm_security": "indirect: direct address to AI/assistant",
"llm_ingestion_guard": "indirect:direct-address"
}
},
{
"id": "indirect:trigger-based",
"label": "indirect: trigger-based instruction for AI",
"pattern": "when\\s+(?:you|the\\s+AI|the\\s+assistant|Claude)\\s+(?:read|see|encounter|process)\\s+this",
"flags": "i",
"aliases": {
"llm_security": "indirect: trigger-based instruction for AI",
"llm_ingestion_guard": "indirect:trigger-based"
}
},
{
"id": "multi-lang:french",
"label": "multi-lang: French \"ignore previous instructions\"",
"pattern": "ignor(?:ez?|er?)\\s+(?:les?\\s+)?instructions?\\s+pr[e\\u00e9]c[e\\u00e9]dentes?",
"flags": "i",
"aliases": {
"llm_security": "multi-lang: French \"ignore previous instructions\"",
"llm_ingestion_guard": "multi-lang:french"
}
},
{
"id": "multi-lang:spanish",
"label": "multi-lang: Spanish \"ignore previous instructions\"",
"pattern": "ignor(?:ar?|e)\\s+(?:las?\\s+)?instrucciones?\\s+anteriores?",
"flags": "i",
"aliases": {
"llm_security": "multi-lang: Spanish \"ignore previous instructions\"",
"llm_ingestion_guard": "multi-lang:spanish"
}
},
{
"id": "multi-lang:german",
"label": "multi-lang: German \"ignore previous instructions\"",
"pattern": "ignorier(?:e|en)?\\s+(?:alle\\s+)?vorherigen?\\s+(?:Anweisungen|Instruktionen)",
"flags": "i",
"aliases": {
"llm_security": "multi-lang: German \"ignore previous instructions\"",
"llm_ingestion_guard": "multi-lang:german"
}
},
{
"id": "markdown:link-ref-comment",
"label": "markdown: suspicious instruction in link-reference comment",
"pattern": "\\[\\/\\/\\]:\\s*#\\s*\\(.*(?:ignore|override|system|instruction|execute)",
"flags": "i",
"aliases": {
"llm_security": "markdown: suspicious instruction in link-reference comment",
"llm_ingestion_guard": "markdown:link-ref-comment"
}
},
{
"id": "data-uri:executable",
"label": "data-uri: executable content type",
"pattern": "data:(?:text\\/html|application\\/javascript|text\\/javascript)[;,]",
"flags": "i",
"aliases": {
"llm_security": "data-uri: executable content type",
"llm_ingestion_guard": "data-uri:executable"
}
},
{
"id": "markdown:link-anchor-injection",
"label": "markdown: injection payload in link anchor text",
"pattern": "\\[[^\\]]*(?:system|ignore|override|exfiltrate|execute)[^\\]]*\\]\\([^)]+\\)",
"flags": "i",
"aliases": {
"llm_security": "markdown: injection payload in link anchor text",
"llm_ingestion_guard": "markdown:link-anchor-injection"
}
},
{
"id": "sub-agent:spawn-dangerous",
"label": "sub-agent: spawn instruction with dangerous capability keywords",
"pattern": "(?:create|spawn|launch|start|run)\\s+(?:an?\\s+)?(?:new\\s+)?(?:sub-?agent|agent|task|worker)\\s+(?:that|to|which|with)\\s+(?:.*?\\s+)?(?:execute|run|delete|remove|send|post|exfiltrate|access|reads?\\s+(?:.*?\\s+)?(?:secret|credential|key|token|\\.env|\\.ssh))",
"flags": "i",
"aliases": {
"llm_security": "sub-agent: spawn instruction with dangerous capability keywords",
"llm_ingestion_guard": "sub-agent:spawn-dangerous"
}
},
{
"id": "sub-agent:delegate-bypass",
"label": "sub-agent: delegation with safety bypass instruction",
"pattern": "(?:delegate|dispatch)\\s+(?:to\\s+)?(?:an?\\s+)?(?:new\\s+)?(?:agent|sub-?agent|task)\\s+.*?(?:bypass|override|ignore|without\\s+(?:review|confirmation|approval))",
"flags": "i",
"aliases": {
"llm_security": "sub-agent: delegation with safety bypass instruction",
"llm_ingestion_guard": "sub-agent:delegate-bypass"
}
},
{
"id": "nl-indirection:fetch-execute",
"label": "nl-indirection: fetch URL and execute",
"pattern": "fetch\\s+(?:this|the|that)\\s+(?:URL|link|endpoint)\\s+and\\s+(?:execute|run|eval)",
"flags": "i",
"aliases": {
"llm_security": "nl-indirection: fetch URL and execute",
"llm_ingestion_guard": "nl-indirection:fetch-execute"
}
},
{
"id": "nl-indirection:send-external",
"label": "nl-indirection: send data to external destination",
"pattern": "send\\s+(?:this|the|that|all)\\s+(?:data|content|output|result|information|file)\\s+to\\s+(?:https?:\\/\\/|ftp:\\/\\/|\\S+@\\S+)",
"flags": "i",
"aliases": {
"llm_security": "nl-indirection: send data to external destination",
"llm_ingestion_guard": "nl-indirection:send-external"
}
},
{
"id": "nl-indirection:read-dotfile",
"label": "nl-indirection: read sensitive dotfile path",
"pattern": "read\\s+(?:the\\s+)?(?:contents?\\s+of\\s+)?(?:~\\/)?\\.(?:ssh|aws|gnupg|env)\\b",
"flags": "i",
"aliases": {
"llm_security": "nl-indirection: read sensitive dotfile path",
"llm_ingestion_guard": "nl-indirection:read-dotfile"
}
},
{
"id": "nl-indirection:extract-exfiltrate",
"label": "nl-indirection: extract credentials and exfiltrate",
"pattern": "(?:extract|copy|dump|exfiltrate)\\s+(?:all\\s+)?(?:the\\s+)?(?:credentials?|secrets?|tokens?|keys?|passwords?)\\s+(?:and|then)\\s+(?:send|post|upload|transmit)",
"flags": "i",
"aliases": {
"llm_security": "nl-indirection: extract credentials and exfiltrate",
"llm_ingestion_guard": "nl-indirection:extract-exfiltrate"
}
},
{
"id": "nl-indirection:download-execute",
"label": "nl-indirection: download and execute",
"pattern": "(?:download|fetch|retrieve)\\s+(?:from\\s+)?\\S+\\s+(?:and|then)\\s+(?:execute|run|eval|pipe\\s+to\\s+(?:bash|sh|node|python))",
"flags": "i",
"aliases": {
"llm_security": "nl-indirection: download and execute",
"llm_ingestion_guard": "nl-indirection:download-execute"
}
}
]
},
{
"id": "hybrid",
"source_export": "HYBRID_PATTERNS",
"severity": null,
"severity_note": "The seed dump supplies this array but NOT the severity the engine assigns to it. Left null rather than inferred from the three sibling arrays. A consumer MUST NOT assume a severity for this family until the producing module is published.",
"count": 8,
"patterns": [
{
"id": "hybrid-p2sql:inject-sql",
"label": "hybrid-p2sql: injection + SQL keywords (prompt-to-SQL attack)",
"pattern": "(?:ignore|override|disregard|forget)[^.]{0,60}(?:SELECT\\s+\\*|DROP\\s+TABLE|UNION\\s+SELECT|DELETE\\s+FROM|INSERT\\s+INTO|UPDATE\\s+\\w+\\s+SET)(?:\\b|(?=\\s|$))",
"flags": "i",
"aliases": {
"llm_security": "hybrid-p2sql: injection + SQL keywords (prompt-to-SQL attack)",
"llm_ingestion_guard": "hybrid-p2sql:inject-sql"
}
},
{
"id": "hybrid-p2sql:sql-inject",
"label": "hybrid-p2sql: SQL operation + injection override keywords",
"pattern": "(?:SELECT\\s+\\*|DROP\\s+TABLE|UNION\\s+SELECT|DELETE\\s+FROM)\\s[^;]{0,80}(?:ignore|override|disregard|bypass)",
"flags": "i",
"aliases": {
"llm_security": "hybrid-p2sql: SQL operation + injection override keywords",
"llm_ingestion_guard": "hybrid-p2sql:sql-inject"
}
},
{
"id": "hybrid-recursive:inject-output",
"label": "hybrid-recursive: instruction to inject into model output",
"pattern": "(?:inject|insert|embed|include)\\s+(?:this|the\\s+following)\\s+(?:into|in)\\s+(?:your|the)\\s+(?:output|response|reply|message|prompt|context)",
"flags": "i",
"aliases": {
"llm_security": "hybrid-recursive: instruction to inject into model output",
"llm_ingestion_guard": "hybrid-recursive:inject-output"
}
},
{
"id": "hybrid-recursive:conditional-response",
"label": "hybrid-recursive: conditional response injection (recursive payload)",
"pattern": "(?:when|if)\\s+(?:the\\s+)?(?:user|human|operator)\\s+(?:asks?|requests?|queries)[^.]{0,60}(?:respond\\s+with|output|reply\\s+with|include)\\s+(?:this|the\\s+following)",
"flags": "i",
"aliases": {
"llm_security": "hybrid-recursive: conditional response injection (recursive payload)",
"llm_ingestion_guard": "hybrid-recursive:conditional-response"
}
},
{
"id": "hybrid-xss:script-tag",
"label": "hybrid-xss: <script> tag in content (agent context XSS)",
"pattern": "<script\\b[^>]*>[\\s\\S]*?<\\/script>",
"flags": "i",
"aliases": {
"llm_security": "hybrid-xss: <script> tag in content (agent context XSS)",
"llm_ingestion_guard": "hybrid-xss:script-tag"
}
},
{
"id": "hybrid-xss:javascript-uri",
"label": "hybrid-xss: javascript: URI scheme (agent context XSS)",
"pattern": "javascript\\s*:",
"flags": "i",
"aliases": {
"llm_security": "hybrid-xss: javascript: URI scheme (agent context XSS)",
"llm_ingestion_guard": "hybrid-xss:javascript-uri"
}
},
{
"id": "hybrid-xss:event-handler",
"label": "hybrid-xss: inline event handler attribute (agent context XSS)",
"pattern": "\\bon(?:error|load|click|mouseover|focus|blur)\\s*=",
"flags": "i",
"aliases": {
"llm_security": "hybrid-xss: inline event handler attribute (agent context XSS)",
"llm_ingestion_guard": "hybrid-xss:event-handler"
}
},
{
"id": "hybrid-xss:iframe-src",
"label": "hybrid-xss: iframe with executable src (agent context XSS)",
"pattern": "<iframe\\b[^>]*src\\s*=\\s*[\"'][^\"']*(?:javascript:|data:text\\/html)",
"flags": "i",
"aliases": {
"llm_security": "hybrid-xss: iframe with executable src (agent context XSS)",
"llm_ingestion_guard": "hybrid-xss:iframe-src"
}
}
]
}
],
"totals": {
"families": 4,
"patterns": 83
}
}
Reference in a new issue View git blame Copy permalink