skills/harvest-rules/SKILL.md
<agentic_guild_skill> <skill_definition> <name>harvest-rules</name> <description>Analyzes branch changes to identify new architectural patterns and map them to the correct documentation files.</description> </skill_definition> <state_machine_directives> 1. NEVER execute more than ONE <step> per response. 2. When you see [PAUSE], you MUST completely stop generating text and wait for the user to reply. 3. Always end your response by summarizing our progress in a conversa
npx skillsauth add jdugarte/agentcore skills/harvest-rulesInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
<agentic_guild_skill> <skill_definition> <name>harvest-rules</name> <description>Analyzes branch changes to identify new architectural patterns and map them to the correct documentation files.</description> </skill_definition>
<state_machine_directives> 1. NEVER execute more than ONE <step> per response. 2. When you see [PAUSE], you MUST completely stop generating text and wait for the user to reply. 3. Always end your response by summarizing our progress in a conversational manner and gently inviting the user to proceed. </state_machine_directives>
<hard_constraints>
NEVER use any tool to execute git commit, git push, or git merge. These commands are STRICTLY FORBIDDEN.
When a commit is appropriate, output a suggested message as a plain-text code block only. The user runs all git commands themselves.
</hard_constraints>
<pre_flight>
<directive>Before executing the workflow, verify the necessary context exists.</directive>
<check>Verify docs/core/SYSTEM_ARCHITECTURE.md and docs/core/SPEC.md exist.</check>
<action>If they are missing, pause our work and gently let the user know we need these files to start. Offer to gracefully initialize the project templates for them. If the user says yes, run sync.sh (or equivalent) if available; otherwise create minimal placeholders from EXPECTED_PROJECT_STRUCTURE. Do NOT hallucinate contents without user confirmation.</action>
</pre_flight>
<phase id="2" name="Knowledge Commit">
<step id="2.1">
<action>
Parse the user's approval. Accept phrasings such as "APPROVE", "Yes", "Write rules", or a list of which candidates to apply.
Write the approved rules directly into the corresponding target files (e.g., `.cursorrules`, `docs/core/SYSTEM_ARCHITECTURE.md`) physically using the `replace_file_content` or `write_to_file` tools.
Once the knowledge transfer is complete, clear `.agenticguild/review_ledger.md` using file tools so the short-term memory is fresh for the next branch. This must be done unconditionally; any ledger entries that did not result in an approved rule are considered implicitly rejected by the user and should be garbage collected.
</action>
<yield>[PAUSE - RULES HARVESTED. SKILL COMPLETE]</yield>
</step>
</phase>
</workflow>
</agentic_guild_skill>data-ai
<agentic_guild_skill> <skill_definition> <name>update-agentic-guild</name> <description>Intelligently synchronizes and updates agentic:guild OS components (skills, rules, templates) from the global repository, using AI to merge changes gracefully. For projects that already have agentic:guild installed, this skill replaces running sync.sh manually.</description> </skill_definition> <state_machine_directives> 1. NEVER execute more than ONE <step> per response. 2. When you se
development
<agentic_guild_skill> <skill_definition> <name>sync-docs</name> <description>Keeps project docs in sync with branch changes and with task memory. Uses two inputs: (1) the branch diff — to infer code/schema-driven doc updates; (2) the active task session file when present — semantically analyzed so domain, decisions, data semantics, and rule-worthy content are pushed to the right docs. Does not dump raw session content; synthesizes and places knowledge by type.</description> </skill_d
tools
<agentic_guild_skill> <skill_definition> <name>status-check</name> <description>Rehydrates project context and acts as the GPS for agentic:guild execution state.</description> </skill_definition> <state_machine_directives> 1. NEVER generate or modify application code during this skill. 2. Your ONLY job is diagnosis and context rehydration. </state_machine_directives> <hard_constraints> NEVER use any tool to execute `git commit`, `git push`, or `git merge`. These co
development
<agentic_guild_skill> <skill_definition> <name>start-task</name> <description>Initiates the process of building a new feature, bugfix, refactor, or chore. Enforces strict QA discovery, implementation planning, and TDD.</description> </skill_definition> <state_machine_directives> 1. NEVER execute more than ONE <step> per response. 2. When you see [PAUSE], you MUST completely stop generating text and wait for the user to reply. 3. Always end your response by summarizing