Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add commands/dtannen/pm/assessgit clone --depth 1 https://github.com/dtannen/pmWhat it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00008 | $0.02603 |
| Opus 5 | $0.00004 | $0.01301 |
| Sonnet 5 | $0.00002 | $0.00521 |
| Haiku 4.5 | $0.00001 | $0.00260 |
Grade A, and why
assess scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 374 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Task Complexity Assessment
Analyze task complexity and provide detailed mode recommendations without executing the task.
Usage
/pm:assess <task-name>
Example: /pm:assess user-authentication
Instructions
You are providing standalone complexity assessment for the PM Adaptive workflow, allowing users to understand task complexity and mode recommendations before execution.
MCP Integration (preferred)
Use MCP to resolve tasks and persist assessment results.
- Full name: mcp__project-manager-mcp__list_tasks
- Full name: mcp__project-manager-mcp__get_task_details
- Full name: mcp__project-manager-mcp__update_task
- Full name: mcp__project-manager-mcp__create_task (if the task does not yet exist)
Recommended flow:
- Lookup task by name via
mcp__project-manager-mcp__list_tasks; if not found, create using...__create_task(name, project_name?, epic_name?, ra_mode?, ra_score?). - Fetch details via
mcp__project-manager-mcp__get_task_detailsfor context and prior RA metadata. - After assessment, persist:
mcp__project-manager-mcp__update_task(ra_score, ra_mode, ra_metadata.assessment, log_entry). - Do not change status; leave execution to
/pm:task-start.
1. Task Validation
- Prefer MCP: resolve by name using
mcp__project-manager-mcp__list_tasks; if missing and appropriate, create with...__create_task. - Fallback: check if
.pm/tasks/$ARGUMENTS.mdexists. - If task definition is incomplete, prompt for missing details.
2. Deploy Assessment Agent
Use the Task tool to deploy the adaptive-assessor agent:
Task:
description: "Detailed complexity assessment for $ARGUMENTS"
subagent_type: "general-purpose"
prompt: |
You are the adaptive complexity assessor providing detailed standalone analysis.
ASSESSMENT TASK: Comprehensive complexity analysis and mode recommendation
Task File: .pm/tasks/$ARGUMENTS.md
DETAILED ASSESSMENT PROTOCOL:
1. TASK ANALYSIS
- Read complete task specification
- Extract estimated hours, domains, integration points
- Identify risk factors and unknowns
- Parse acceptance criteria complexity
2. COMPLEXITY SCORING
Apply detailed scoring matrix:
Base Scores (choose highest applicable):
- Simple file change: 1 point
- Multi-file change: 2 points
- New feature: 3 points
- Refactoring: 4 points
- Architecture change: 5 points
Modifiers (cumulative):
- External API integration: +2
- Database schema changes: +3
- Breaking changes: +3
- Security-critical: +2
- Performance-critical: +2
- Multi-domain (per additional domain): +1
- Estimated >8 hours: +2
- Estimated >16 hours: +3
Keyword Analysis:
- "refactor": +3
- "integrate" / "integration": +2
- "migrate" / "migration": +3
- "architect" / "architecture": +4
- "optimize" / "performance": +2
- "authentication" / "security": +2
- "payment" / "billing": +3
- "real-time" / "websocket": +2
3. CODEBASE CONTEXT
- Scan mentioned files and directories
- Identify affected system boundaries
- Count integration points
- Assess existing code complexity
- Evaluate testing requirements
4. RISK ASSESSMENT
- Identify assumptions that may be wrong
- Flag potential integration issues
- Note areas with high uncertainty
- Assess impact of potential mistakes
5. MODE RECOMMENDATION
Based on final score:
- 1-3: Simple Mode
- 4-6: Standard Mode
- 7-8: RA-Light Mode
- 9-10: RA-Full Mode
DELIVERABLE FORMAT:
Provide comprehensive assessment using this structure:
# COMPLEXITY ASSESSMENT: $ARGUMENTS
## Quick Summary
**Score**: X/10 | **Mode**: [Recommended] | **Confidence**: [High/Medium/Low]
## Detailed Breakdown
### Base Complexity Analysis
**Primary Task Type**: [classification] → X points
**Rationale**: [why this classification]
### Complexity Modifiers Applied
- [Modifier 1]: +X points - [specific reason]
- [Modifier 2]: +X points - [specific reason]
[list all applicable modifiers]
### Keyword Impact
**Keywords Detected**: [list relevant keywords]
**Keyword Score**: +X points total
### Codebase Context
**Files Likely Affected**: ~X files
**Domains Involved**: [list domains]
**Integration Points**: [list external systems/APIs]
**System Boundaries Crossed**: X boundaries
### Risk Factors
**High Risk Areas**:
- [Risk 1]: [description and impact]
- [Risk 2]: [description and impact]
**Uncertainty Areas**:
- [Unknown 1]: [what's unclear]
- [Unknown 2]: [what's unclear]
### Final Calculation
**Base Score**: X
**Modifiers**: +X
**Keyword Bonus**: +X
**Context Adjustment**: +/-X
**Total Score**: X/10
## Mode Recommendation: [MODE NAME]
### Why This Mode?
[Detailed rationale for mode selection]
### What This Mode Provides:
[Description of workflow, rigor level, verification]
### Expected Workflow:
[Step-by-step process description]
### Estimated Timeline:
- **Assessment**: X minutes
- **Planning**: X minutes
- **Implementation**: X hours
- **Verification**: X minutes/hours
- **Total**: X hours
## Alternative Considerations
### If Score Were Higher/Lower:
- **One mode up**: [when to consider escalating]
- **One mode down**: [when simpler might work]
### Override Scenarios:
**Consider Simple Mode if**:
- [conditions that might reduce complexity]
**Consider RA-Full if**:
- [conditions that might increase risk]
## Pre-Implementation Checklist
### Before Starting:
- [ ] [Preparation item 1]
- [ ] [Preparation item 2]
### Red Flags to Watch:
- [Warning sign 1]: Escalate to higher mode
- [Warning sign 2]: Stop and reassess
## Learning Notes
**Assessment Confidence**: [High/Medium/Low]
**Key Uncertainty**: [biggest unknown factor]
**Track This**: [what to measure for future learning]
---
Ready to execute with: `/pm:start $ARGUMENTS`
Override mode with: `/pm:start $ARGUMENTS --mode=[mode]`
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 2d ago First seen · 374 lines · 8 tokens per session scan A e17869ba9bd3
assess is a command published in the GitHub repository dtannen/pm (0 stars, last pushed 10mo ago), licensed MIT. It adds 8 tokens to every session and 2,603 once invoked, about $0.0000 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other commands, from other repositories
git
Git operations with intelligent commit messages and workflow optimization.
checklist
Generate a custom checklist for the current feature based on user requirements.
clarify
Identify underspecified areas in the current feature spec by asking up to 5 highly targeted clarification questions and encoding answers back into the spec.
specify
Create or update the feature specification from a natural language feature description.
analyze
Perform a non-destructive cross-artifact consistency and quality analysis across spec.md, plan.md, and tasks.md after task generation.
converge
Assess the current codebase against the feature's spec, plan, and tasks, then append any remaining unbuilt work as new tasks to tasks.md so implement can complete it.