BC agent-memory-cleanup
Clean and audit long-term agent memory files such as user.md, memory.md, memories.md, profile.md, preferences.md, and agent_memory.md. Use when the user explicitly asks to clean, prune, sanitize, deduplicate, or review agent memory files, or when a memory write/update fails because the memory file is too long, full, over budget, rejected, duplicated, conflicted, stale, task-specific, or contains suspected secrets. Do not trigger for ordinary project docs, task notes, logs, code review, README edits, or general file cleanup. Applies edits only after explicit approval and creates recoverable backups.
As a process C 62/100 · Has gaps — weak spots: result and completion, inputs and preconditions, progress reporting
What is at stake
Medium-severity findings: the skill is probably honest, but read what alarmed the scanner.
Below is the worst case for this category. The finding here is medium: the guard saw a sign, not a proof.
The files contain someone else's key or token. If it is live, your agent will call third-party services under a stranger's identity; if it was revoked, the skill's scripts simply fail. Such a key often arrives with the author's whole workspace, personal data included.
The key is visible to everyone who downloaded the skill and has likely been copied by catalog-scanning bots already. Revoke it now, check bills and access logs, then reissue.
How to improve
- A spec.yaml with trigger phrases and assertions — a behaviour contract for CI; `skilltest init` writes a template.
Guard findings · 7
✓ No critical or high findings
Medium and low: 7
-
medium Secrets in code
secret-labelled-tokentest-fixtures/secret-memory.md:4Labelled token / key literal (vendor format unknown — verify it is not a live credential)- Temporary credential for the staging system: api_key=abc1…lue
-
low Secrets in code
secret-high-entropy-tokenscripts/run_tests.py:40High-entropy token-like string (may be an id, hash or a credential) (test fixture / example file; quoted — discussed, not commanded)assert_true("eyJh…CJ9" not in output, "bearer token leaked in JSON output")fixturequoted -
low Secrets in code
secret-high-entropy-tokenscripts/run_tests.py:42High-entropy token-like string (may be an id, hash or a credential) (test fixture / example file; quoted — discussed, not commanded)assert_true("pypi…zg5" not in output, "pypi token leaked in JSON output")fixturequoted -
low Secrets in code
secret-high-entropy-tokenscripts/run_tests.py:94High-entropy token-like string (may be an id, hash or a credential) (test fixture / example file; quoted — discussed, not commanded)"eyJh…CJ9",
fixturequoted -
low Secrets in code
secret-high-entropy-tokenscripts/run_tests.py:96High-entropy token-like string (may be an id, hash or a credential) (test fixture / example file; quoted — discussed, not commanded)"pypi…zg5",
fixturequoted -
low Secrets in code
secret-password-literaltest-fixtures/secret-memory.md:4Hard-coded password / key literal (may be an example)- Temporary credential for the staging system: api_key=abc1…lue
-
low Secrets in code
secret-high-entropy-tokentest-fixtures/secret-memory.md:7High-entropy token-like string (may be an id, hash or a credential)- PyPI token from packaging notes: pypi…zg5
Files scanned: 16. Evidence is masked. Grey chips explain why severity was lowered.
Against the Agent Skills spec
✓ No remarks against the Agent Skills spec
Process rating: all ten parameters 62/100
- 0Result and completion. Does not say what the result is
- 0Inputs and preconditions. Does not say what the process needs to start
- 0Progress reporting. Says nothing while it works
- 60Tools and files. Uses tools (python) that frontmatter does not declare
- 70When it triggers. States when to use, but not when not to
- 100Steps. 25 steps
- 100Failures and branches. 2 branches, has a failure section
- 100Consistency. Name and required fields are in place
- 100Execution cost. Instruction body is 750 tokens
- 100Running it twice. Mutating operations check current state
- medium Safety rules and hard prohibitions inside a skill: they belong in the system prompt, here they protect nothing
- low No test case covers injection arriving through data
Everything here is measured from the skill text rather than judged by a model, so the numbers are checkable. A parameter weighs more when it is a more common reason for the process to stall.
Quality signals
- +5Description has no quoted example phrases that should trigger the skill
- +4Description does not say when NOT to use the skill (false activations)
- +3Output format is not stated: the model decides each time
- -31 of 2 scripts are never mentioned in SKILL.md
- +2Single-language instructions
- +3Description length 605: enough signal without eating the budget
- +4Structure: 6 headings
- +3Step-by-step instructions: 25 items
- +4Has examples (4 code blocks)
- +4Reference files are cited in the instructions (3 of 3)
- +1License stated
Quality base 70; lint remarks subtract, signals add up to 100. Result: 86.