audit-log-of-tool-calls
Skill0 starsChanged 18 days agoMITView on GitHub
Log every call with arguments and outcome so an agent's actions can be reconstructed. Use when the server can change state or reach anything outside its own directory
In this repository49 other files it gives its agents.
- argument-construction-from-artifactsSkill
- bounded-output-designSkill
- capability-negotiation-checkSkill
- changelog-and-versioningSkill
- concurrency-and-ordering-checkSkill
- context-window-managementSkill
- credential-isolationSkill
- data-retention-policySkill
- description-as-contractSkill
- destructive-action-confirmationSkill
- destructive-tool-gatingSkill
- deterministic-command-choicesSkill
- done-when-criteriaSkill
- error-shape-for-agentsSkill
- fixture-driven-transcript-testsSkill
- human-in-the-loop-escalationSkill
- idempotency-annotationSkill
- input-schema-strictnessSkill
- least-privilege-tool-grantsSkill
- lifecycle-test-initialize-then-toolsSkill
- local-vs-remote-boundarySkill
- one-procedure-per-skillSkill
- pagination-and-limit-disciplineSkill
- parallel-tool-call-safetySkill
- protocol-conformance-testSkill
- protocol-version-handshakeSkill
- publishing-a-pack-locallySkill
- rate-limit-respectSkill
- reference-canonicalisationSkill
- replay-transcript-for-regressionSkill
- restart-and-reload-semanticsSkill
- scope-guard-before-executionSkill
- search-then-fetch-patternSkill
- secrets-via-environmentSkill
- server-discovery-and-trustSkill
- signal-and-pitfall-sectionsSkill
- skill-frontmatter-contractSkill
- stdio-client-configSkill
- stdout-purity-checkSkill
- third-party-server-sandboxingSkill
- tool-call-budgetingSkill
- tool-call-error-pathsSkill
- tool-error-without-retry-stormSkill
- tools-list-schema-validationSkill
- tool-surface-minimisationSkill
- tool-verb-namingSkill
- transport-choiceSkill
- verify-tool-output-before-reportingSkill
- when-to-use-and-avoidSkill
The file
We split this skill into the rules it gives an agent
2 rules. About 0 are common, 2 are distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈2
Distinctive · estimate
This author's own choices. Worth a look.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 25
Verify a session can be reconstructed from the log alone
- Distinctive: line 63
no credential value appears in any log line
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Audit Log of Tool Calls
Log every call with arguments and outcome so an agent's actions can be reconstructed
Use when: the server can change state or reach anything outside its own directory
Avoid when: the server is read-only over public content and no audit is required
Procedure
- Decide the log destination: stderr, a file or both, never stdout
- Log timestamp, tool name, argument names and outcome
- Redact argument values that are credentials
- Log the failure reason for refused and errored calls
- Rotate or cap the log so it cannot fill the disk
- Verify a session can be reconstructed from the log alone
Signals
- each call appears exactly once in the log
- argument values that are secrets are redacted
- refusals and errors are logged with reasons
- the log never appears on stdout
Patterns
1. Structured line
{"ts":...,"tool":"get_skill","args":["slug"],"ok":true}2. Redaction
args values logged as [redacted] when the key matches token|key|secret3. Log size cap
rotate at 10 MB, keep three filesPitfalls
- Logging whole argument objects including credentials
- Writing the audit log to stdout and breaking the protocol
- Logging failures without the reason, making the entry useless
Done when
- a session can be reconstructed from the log
- no credential value appears in any log line
References
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

