idempotency-annotation
Skill0 starsChanged 14 days agoMITView on GitHub
Mark which tools change state so a retry after a timeout cannot duplicate an action. Use when a tool writes to disk, sends a request or mutates a remote record
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
- 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 16
Avoid when: every tool is a pure read, which is the safest design for a skill server
- Distinctive: line 25
Test a repeated identical write and confirm no second effect
≈ “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
Idempotency Annotation
Mark which tools change state so a retry after a timeout cannot duplicate an action
Use when: a tool writes to disk, sends a request or mutates a remote record
Avoid when: every tool is a pure read, which is the safest design for a skill server
Procedure
- Classify each tool as read, write or destructive
- Say which class it is in the tool description, in plain words
- Design writes to be idempotent where possible, keyed by content or id
- Refuse retries of destructive tools without an explicit identifier
- Log the identifier so a duplicate can be recognised later
- Test a repeated identical write and confirm no second effect
Signals
- a repeated write with the same key produces one effect
- the description states whether the call mutates state
- destructive tools require an explicit target argument
- logs show the key that made the operation idempotent
Patterns
1. Class in words
read-only: returns indexed skills, changes nothing2. Content key
idempotency key = sha256(payload)3. Refusal shape
refusing repeat delete without an explicit idPitfalls
- Leaving idempotency implicit and letting an agent retry a write
- Making every write destructive and unguarded because the tool is internal
- Hashing a payload that includes a timestamp, which defeats the key
Done when
- each tool states whether it mutates state
- a repeated identical write is proven to have one effect
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.

