verify-tool-output-before-reporting
Skill0 starsChanged 15 days agoMITView on GitHub
Treat tool output as an observation to check, never as a conclusion to publish. Use when a tool result will be quoted in a report, ticket or decision
In this repository49 other files it gives its agents.
- argument-construction-from-artifactsSkill
- audit-log-of-tool-callsSkill
- 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
- when-to-use-and-avoidSkill
The file
We split this skill into the rules it gives an agent
3 rules. About 0 are common, 3 are distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈3
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 20
Read what the tool claims and what it actually returned
- Distinctive: line 21
Check the claim against the raw artifact, not against the formatted summary
- Distinctive: line 62
every quoted result is traceable to an artifact and a rerun
≈ “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
Verify Tool Output Before Reporting
Treat tool output as an observation to check, never as a conclusion to publish
Use when: a tool result will be quoted in a report, ticket or decision
Avoid when: the output is a step toward a later measurement that will be verified anyway
Procedure
- Read what the tool claims and what it actually returned
- Check the claim against the raw artifact, not against the formatted summary
- Re-run the call once to confirm the output is stable
- Confirm the output's scope matches the question you asked
- Note the tool version and arguments beside the quoted result
- Report only what you can re-derive from the artifact
Signals
- the quoted text appears in the artifact at the cited location
- a rerun produces the same result
- the answer addresses the question that was actually asked
- the tool version and arguments appear with the quote
Patterns
1. Claim versus artifact
summary says 50 skills; pack.json says skill_count 502. Rerun check
same call twice, identical output3. Provenance
skillmcp 1.0.0, search_skills("cache")Pitfalls
- Quoting a formatted summary as if it were the underlying data
- Publishing a first-run result that changes on a second run
- Reporting an answer to a different question than the one asked
Done when
- every quoted result is traceable to an artifact and a rerun
- the tool version and arguments accompany the quote
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.

