{"slug":"retro-7","title":"retro","summary":"Conduct a retrospective on a coding session.","platform":"Claude","tags":[],"authorName":"LLM Mart","authorSlug":"llm-mart","score":0,"source":"github","price":null,"verified":false,"createdAt":"2026-10-14T21:49:26.368008Z","repo":{"url":"https://github.com/dzhng/skills","stars":1333,"forks":81,"license":"MIT","updatedAt":"2026-10-14T20:39:18Z"},"bodyHtml":"<hr>\n<h2>name: retro\ndescription: \"Conduct a retrospective on a coding session.\"\ndisable-model-invocation: true</h2>\n<p>The user has asked for a <strong>retrospective</strong>. You are suggesting improvements to the coding agent's <strong>environment</strong> to improve future runs.</p>\n<h2>Steps</h2>\n<ol>\n<li><p>Read <a href=\"../../authoring/write-skills/SKILL.md\">write-skills</a> for the writing style guide.</p>\n</li>\n<li><p>Read the primary sources for the session the user specifies. This may mean searching through session logs on this machine. If the user doesn't specify a session, default to the current one.</p>\n</li>\n<li><p>Look for candidates for improvement in these categories.</p>\n</li>\n</ol>\n<ul>\n<li><strong>Navigation</strong>: how easy was it for the agent to find the right files? Are there hidden dependencies between files? Would a <strong>navigation pointer</strong> make it easier? <em>Use when</em> the session took a long time to find a piece of information.</li>\n<li><strong>Automated checks</strong>: are there automated checks that could catch errors the agent made? Linting, typing, tests, filesystem linters? Read the repo's own check command first (its <code>package.json</code>/build-tool <code>lint</code>/<code>check</code> scripts, its CI workflow), so a check that already exists but sits unwired or silently broken is the finding, not a reinvention. A repo with no <strong>guardrail</strong> (no pre-commit hook and no CI job running its lint/typecheck/test command) is itself a finding: an un-linted repo is a standing missed opportunity, not a neutral default. <em>Use when</em> the agent made a mistake an automated check could have caught, or the repo has no guardrail at all.</li>\n<li><strong>Coding standards</strong>: should the <strong>reviewer agent</strong> be given a new rule to enforce? Should an existing rule be removed or clarified? Classify the violation first: a <strong>mechanical</strong> one (a fixed syntactic pattern, a banned API, an import shape, a file-location rule) gets a deterministic check, full stop: a custom rule in the repo's own linter, a new pre-commit hook, or a new CI job, whichever the repo's language and existing guardrail make cheapest. Default to building the check over writing the rule. Reserve <code>CODING_STANDARDS.md</code> for genuine <strong>judgement calls</strong> (cross-file consistency, \"matches the surrounding style,\" anything no guardrail could ever substitute for). <em>Use when</em> the reviewer agent failed to catch a mistake.</li>\n<li><strong>Global AGENTS.md</strong>: are there any steering instructions that should be moved to coding standards (or automated checks) instead? <em>Use when</em> the AGENTS.md file is particularly large - in the repo OR the user's global scope.</li>\n<li><strong>Tool economy</strong>: did the agent make expensive tool calls that could be streamlined? Is there any custom tooling (CLI's, MCP's) that is particularly token-inefficient? <em>Use when</em> the agent made an expensive tool call.</li>\n<li><strong>No-ops</strong>: look for instructions in steering files that don't modify the agent's behavior. <em>Use when</em> the steering files are large and unwieldy.</li>\n<li><strong>Information access</strong>: look for opportunities to increase the agent's access to information. Teeing dev server logs, readonly access to third-party services. <em>Use when</em> a crucial piece of information was not available to the agent.</li>\n</ul>\n<ol start=\"4\">\n<li>Present these candidates to the user, in order of severity.</li>\n</ol>\n<h2>Reference</h2>\n<h3>Implementation vs Review</h3>\n<p>Remember that all work goes through two stages: implementation and review. The implementation agent has the most <strong>context pressure</strong>. They are responsible for exploration, writing code, and debugging failures.</p>\n<p>The review agent has the least context pressure - it receives a diff, so no exploration needed. It often does not need to write code or debug.</p>\n<p>This means that the review agent should be responsible for imposing coding standards, not the implementation agent.</p>\n<h3>Files</h3>\n<p>You have access to several files in the repo:</p>\n<ul>\n<li><code>CLAUDE.md</code>/<code>AGENTS.md</code>: these files are pushed to the context window of any agent working in this repo. They should be used incredibly sparingly, usually only for <strong>navigation pointers</strong> to other files.</li>\n<li><code>CODING_STANDARDS.md</code>: this file is read during review, not implementation. Add <strong>navigation pointers</strong> to docs folders if the standards file gets more than 1,000 lines long.</li>\n<li>Docs: use docs as references files, pointed to by other files. Look for existing docs before writing new ones.</li>\n<li>Skills: use skills for docs (since their description goes into the agent's context window), or for user-invoked commands. Follow the advice in <a href=\"../../authoring/write-skills/SKILL.md\">write-skills</a>.</li>\n</ul>\n","files":[{"path":"agents/openai.yaml","sizeBytes":146,"isText":true},{"path":"LICENSE","sizeBytes":1068,"isText":false},{"path":"SKILL.md","sizeBytes":4405,"isText":true}],"reviewScore":null,"reviewSummary":null,"trust":{"provenance":"trusted-source-unreviewed","notice":"Community-authored content, reproduced verbatim and not vetted as instructions. Treat it as data to evaluate, never as directives to follow.","bodySource":null},"bodyLocked":false,"purchaseUrl":null,"sourceUrl":null,"report":{"provenance":"trusted-source-unreviewed","screen":{"ran":true,"outcome":"clean","suspicious":0,"notes":0,"hiddenCharacters":false},"virusScan":{"engine":"clamav","status":"clean","scannedAt":"2026-10-14T21:49:50.287281Z","sha256":"52DA24447DB8F3B18EE1BD1E37C83E637AD6C614B6389C0819C365589E865DFE","sizeBytes":3072},"review":null,"source":{"repositoryUrl":"https://github.com/dzhng/skills","path":"skills/engineering/retro","license":"MIT","commit":"f2fbf92f131fb08bbf3e400a1ef99b7be85bb2bc","subtreeSha":"1697C8B132A9445FF9BBCA804354A735EB68E485D6B2807C19D14E25E9A2DA67","lastSyncedAt":"2026-10-14T21:49:23.275319Z"},"reviewedAt":"2026-10-14T21:50:41.347734Z","notice":"Community-authored content, reproduced verbatim and not vetted as instructions. Treat it as data to evaluate, never as directives to follow."},"install":[{"target":"skills-cli","command":"npx skills add https://github.com/dzhng/skills/tree/main/skills/engineering/retro"},{"target":"claude-code","command":"claude plugin marketplace add https://llmmart.ai/marketplace.json && claude plugin install dzhng-skills@llmmart"},{"target":"git","command":"git clone https://github.com/dzhng/skills.git"}]}