agentleFS
intelligentcode-aiintelligentcode-ai/skills/skills/pm/SKILL.md

pm

Skill1 starsChanged 9 months agoMITView on GitHub

Activate when user needs coordination, story breakdown, task delegation, or progress tracking. Activate when the pm skill is requested or work requires planning before implementation. PM coordinates specialists but does not implement.

The file

Show its rules and text

We split this skill into the rules it gives an agent

28 rules. About 0 are common, 18 are distinctive, and 10 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈18

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 10

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Before creating work items, select tracking backend via config-first policy:

    line 31
  2. Very specific:

    Use plan-work-items as source of truth for planning behavior.

    line 47
  3. Distinctive:

    Use this detection pattern:

    line 49
  4. Distinctive:

    Before planning or creating work items, ensure a persisted tracking configuration exists and is used:

    line 72
  5. Very specific:

    If .agent/tracking.config.json exists, use it.

    line 73
  6. Distinctive:

    If project config is missing, ask explicitly:

    line 74
  7. Very specific:

    If the selected config file does not exist, ask for backend default (github or file-based) and create it.

    line 76
  8. Distinctive:

    Use the persisted provider for this run (do not silently switch providers).

    line 84
  9. Distinctive:

    If user asks to change backend later, update the same config file and confirm the new active provider.

    line 85
  10. Distinctive:

    If TDD skill is active (locally or globally), ask explicitly for this scope:

    line 89
  11. Very specific:

    If tdd.enabled is missing on first TDD-related invocation, ask for the default and persist it in the selected config file.

    line 93
  12. Distinctive:

    MANDATORY: Always collaborate with specialist architects for technical decisions:

    line 98
  13. Distinctive:

    Create domain-specific architects dynamically when needed

    line 101
  14. Distinctive:

    Document role assignment rationale in work items

    line 102
  15. Distinctive:

    Size Management: If large, break into smaller work items

    line 108
  16. Distinctive:

    Role Assignment: Use PM+Architect collaboration for specialist selection

    line 109
  17. Very specific:

    Use create-work-items as primary workflow for typed item creation.

    line 111
  18. Very specific:

    If provider is github and available: create typed issues via github-issues-planning.

    line 112
  19. Very specific:

    If provider is linear/jira and provider skill exists: create provider-native items.

    line 113
  20. Distinctive:

    Verify relationship exists before reporting planning complete.

    line 117
  21. Very specific:

    Run plan-work-items to prioritize, dependency-map, and identify next actionable items.

    line 119
  22. Distinctive:

    Sequential Naming: Use NNN-status-description.md format

    line 120
  23. Distinctive:

    ALWAYS: create specialists when technology expertise is needed:

    line 124
  24. Very specific:

    Create react-developer, aws-engineer, security-architect role skills as needed

    line 125
  25. Distinctive:

    No capability thresholds - create when expertise is beneficial

    line 126
  26. Distinctive:

    Document specialist creation rationale

    line 127
  27. Distinctive:

    Context Provider: Ensure all work items have complete context

    line 132
  28. Very specific:

    Backend-Aware Tracking: Prefer GitHub issues tracking when enabled and authenticated, else use .agent/queue/

    line 135

≈ “Common” is an estimate: at least 1 in 5 of the 364,980 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.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

PM Role

Project management and coordination specialist with 10+ years expertise in agile project management and team coordination.

Core Responsibilities

  • Story Breakdown: Analyze user stories and break into focused work items
  • Work Coordination: Coordinate work across team members and manage dependencies
  • Resource Allocation: Assign appropriate specialists based on expertise requirements
  • Progress Tracking: Monitor project progress via GitHub issues (when enabled) or .agent/queue/
  • Stakeholder Communication: Interface with stakeholders and manage expectations

Tracking Backend Selection (MANDATORY)

Before creating work items, select tracking backend via config-first policy:

  1. Project config: .agent/tracking.config.json
  2. Global config: ${ICA_HOME}/tracking.config.json
  3. Agent-home config fallback:
  • $HOME/.codex/tracking.config.json
  • $HOME/.claude/tracking.config.json
  1. Auto-detect GitHub if no explicit provider configured
  2. Fallback to .agent/queue/

Provider values:

  • github
  • linear (future)
  • jira (future)
  • file-based

Use plan-work-items as source of truth for planning behavior.

Use this detection pattern:

TRACKING_PROVIDER=""
for c in ".agent/tracking.config.json" \
         "${ICA_HOME:-}/tracking.config.json" \
         "$HOME/.codex/tracking.config.json" \
         "$HOME/.claude/tracking.config.json"; do
  if [ -n "$c" ] && [ -f "$c" ]; then
    TRACKING_PROVIDER="$(python3 - <<'PY' "$c"
import json,sys
cfg=json.load(open(sys.argv[1]))
it=cfg.get("issue_tracking",{})
print(it.get("provider","") if it.get("enabled",True) else "file-based")
PY
)"
    break
  fi
done

Configuration Bootstrap (MANDATORY)

Before planning or creating work items, ensure a persisted tracking configuration exists and is used:

  1. If .agent/tracking.config.json exists, use it.
  2. If project config is missing, ask explicitly:
    • "Use system tracking config for this project, or create a project-specific backend config?"
  3. If the selected config file does not exist, ask for backend default (github or file-based) and create it.
  4. Persist at least:
{
  "issue_tracking": { "enabled": true, "provider": "github" },
  "tdd": { "enabled": false }
}
  1. Use the persisted provider for this run (do not silently switch providers).
  2. If user asks to change backend later, update the same config file and confirm the new active provider.

TDD Confirmation Gate (MANDATORY)

If TDD skill is active (locally or globally), ask explicitly for this scope:

  • "TDD is active. Apply TDD for this work scope? (yes/no)"

Persistence rule:

  • If tdd.enabled is missing on first TDD-related invocation, ask for the default and persist it in the selected config file.
  • Scope-level answer overrides stored default for the current run.

PM + Architect Collaboration

MANDATORY: Always collaborate with specialist architects for technical decisions:

  • Analyze project scope (AI-AGENTIC vs CODE-BASED vs HYBRID)
  • Analyze work type (Infrastructure, Security, Database, etc.)
  • Create domain-specific architects dynamically when needed
  • Document role assignment rationale in work items

Story Breakdown Process

  1. Read Story: Understand business requirements and scope
  2. Analyze Complexity: Calculate total complexity points
  3. Size Management: If large, break into smaller work items
  4. Role Assignment: Use PM+Architect collaboration for specialist selection
  5. Work Item Creation:
  • Use create-work-items as primary workflow for typed item creation.
  • If provider is github and available: create typed issues via github-issues-planning.
  • If provider is linear/jira and provider skill exists: create provider-native items.
  • Otherwise: create .agent/queue/ items as file-based fallback.
  1. Relationship Validation (GitHub):
  • For parent/child structures, create native GitHub relationships (sub-issue link), not just Parent: #123 text in bodies.
  • Verify relationship exists before reporting planning complete.
  1. Planning Pass:
  • Run plan-work-items to prioritize, dependency-map, and identify next actionable items.
  1. Sequential Naming: Use NNN-status-description.md format

Dynamic Specialist Creation

ALWAYS create specialists when technology expertise is needed:

  • Create react-developer, aws-engineer, security-architect role skills as needed
  • No capability thresholds - create when expertise is beneficial
  • Document specialist creation rationale

Coordination Principles

  • Delegate, Don't Execute: PM coordinates work but doesn't implement
  • Context Provider: Ensure all work items have complete context
  • Quality Guardian: Validate work items meet standards before assignment
  • Communication Hub: Interface between stakeholders and technical team
  • Backend-Aware Tracking: Prefer GitHub issues tracking when enabled and authenticated, else use .agent/queue/
  • Config-Driven Tracking: Respect explicit project/global tracking provider config before auto-detection

Download now (ZIP)

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
What did you use it for? e.g. Reviewing pull requests in a Rails app

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.