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/rand/mnemosyne/feature-tasksgit clone --depth 1 https://github.com/rand/mnemosyneWhat 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 | $0.00014 | $0.02505 |
| Opus 5 | $0.00007 | $0.01252 |
| Sonnet 5 | $0.00003 | $0.00501 |
| Haiku 4.5 | $0.00001 | $0.00250 |
Grade C, and why
feature-tasks scanned grade C with 2 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.
Downloads and executes remote codehighSupply chain
curl | sh runs whatever the server returns today, which is not necessarily what it returned when this was reviewed.
- If Beads not available: "Error: Beads CLI not installed. Install:\n npm install -g @beads/bd\n OR: curl -fsSL https://raw.githubusercontent.com/steveyegge/beads/main/scripts/install.sh | bash\n Then run: bd init" Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
- If Beads not available: "Error: Beads CLI not installed. Install:\n npm install -g @beads/bd\n OR: curl -fsSL https://raw.githubusercontent.com/steveyegge/beads/main/scripts/install.sh | bash\n Then run: bd init" How it starts
The opening of the file, as written. The whole thing — 269 lines — stays where its author put it; the contents beside it link to each section on GitHub.
I will help you convert a feature implementation plan into concrete, trackable Beads tasks with proper dependencies and estimates.
Usage:
/feature-tasks <feature-id>- Create tasks from plan/feature-tasks --show <feature-id>- Display existing tasks/feature-tasks --sync <feature-id>- Sync plan changes to tasks
Instructions for me:
-
Load implementation plan:
- Read
.mnemosyne/artifacts/plans/<feature-id>-plan.md - If not found: "Error: Plan '' not found. Use /feature-plan first."
- Parse YAML frontmatter to get plan version, memory_id, feature_id
- Extract implementation order, parallel streams, dependencies
- Read
-
Check for existing tasks:
- Query Beads for tasks with tag
feature:<feature-id> - If exists:
- If
--show: Display task summary and exit - If
--sync: Update tasks based on plan changes - Otherwise: Ask if user wants to view, sync, or recreate
- If
- Query Beads for tasks with tag
-
Analyze plan structure: From plan.md, extract:
- Implementation Order: Ordered list of steps
- Parallel Streams: Independent workstreams
- Dependencies: Task X depends on Task Y relationships
- Integration Points (Typed Holes): Interfaces between components
- Testing Requirements: Test types for each component
-
Generate Beads task structure: For each plan item:
a) Task Title:
- Format:
[<feature-id>] <step-name>: <brief-description> - Example:
[jwt-auth] Implement JWT Generation: RS256 signing with refresh tokens - Max 100 characters
b) Task Description:
## Goal <What needs to be done> ## Technical Details <From plan: approach, data models, APIs> ## Acceptance Criteria - [ ] <criterion 1 from plan> - [ ] <criterion 2 from plan> - [ ] Tests: <test requirements> ## Dependencies - Requires: <task-id> (<task-name>) - Blocks: <task-id> (<task-name>) ## Integration Points - <Typed hole to implement or consume> ## Estimated Effort <hours or days based on plan> ## Related Files - <files from plan> ## Notes <Any additional context from plan>c) Priority:
- P0 (critical): On critical path, blocks multiple tasks
- P1 (high): Important but not blocking
- P2 (medium): Nice-to-have, can defer
- P3 (low): Future enhancement
Determine priority from:
- Critical path items: P0
- Parallel stream items: P1
- Polish/refinement: P2
- Future work: P3
d) Effort Estimate:
- Parse from plan's implementation order
- Default: 4 hours for typical tasks
- Adjust based on complexity indicators:
- "Foundation" or "Core": 4-8 hours
- "Integration": 2-4 hours
- "Polish" or "Refinement": 1-2 hours
- "Testing": 2-4 hours
e) Type:
- feature: New functionality
- refactor: Code improvement
- test: Test addition
- doc: Documentation
- bug: Fix (if plan mentions fixing issues)
f) Tags:
feature:<feature-id>plan:<plan-version><category>(from plan: auth, api, database, etc.)
- Format:
-
Handle dependencies:
- Parse "Dependencies" section from plan
- For "Task X depends on Task Y":
- Create Task Y first (if not exists)
- Add
--blocked-by <task-y-id>to Task X
- For parallel streams:
- No blocking relationships within stream
- May depend on foundation tasks
- For critical path:
- Each task blocks next task
-
Create tasks in Beads (auto-synced):
# Foundation task (no dependencies) - Hash ID assigned bd create "[jwt-auth] Database Schema: Create users table with auth fields" \ -d "$(cat task_desc.md)" \ -t feature -p 0 -l "feature:jwt-auth,plan:1.0.0,database" \ --json # Returns: {"id": "bd-a1b2", ...} # Task with dependency (using hash ID) bd create "[jwt-auth] JWT Generation: Implement RS256 signing" \ -d "$(cat task_desc.md)" \ -t feature -p 0 -l "feature:jwt-auth,plan:1.0.0,auth" \ --json # Returns: {"id": "bd-c3d4", ...} # Add dependency using bd dep bd dep add bd-c3d4 bd-a1b2 --type blocks # Parallel stream task bd create "[jwt-auth] Middleware: Token validation for protected routes" \ -d "$(cat task_desc.md)" \ -t feature -p 1 -l "feature:jwt-auth,plan:1.0.0,middleware" \ --json # Returns: {"id": "bd-e5f6", ...} bd dep add bd-e5f6 bd-c3d4 --type blocks # All tasks auto-synced to .beads/issues.jsonl (5-second debounce)
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 · 269 lines · 14 tokens per session scan C 98c97663cc4f
feature-tasks is a command published in the GitHub repository rand/mnemosyne (84 stars, last pushed 9mo ago), licensed MIT. It adds 14 tokens to every session and 2,505 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it C with 2 findings (downloads and executes remote code, makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other commands, from other repositories
hatch3r-bug-plan
Diagnose a complex incident -- reproduce the symptom, rank root-cause hypotheses, design the fix path, and emit regression coverage items as a board-ready investigation.
hatch3r-debug
Standalone debug-and-fix workflow — add strategic debug logging, collect runtime logs from the user, perform root cause analysis, implement the fix, and clean up all debug artifacts.
hatch3r-feature-plan
Design a new capability -- draft user stories, acceptance criteria, data model, API surface, and sub-issue breakdown as an epic-shaped todo.md for greenfield features.
story-6.1.4
Story ID: 6.1.4 Epic: Epic-6.1 - Agent Identity System Wave: Wave 1 (Foundation) Status: 📋 Ready to Start Priority: 🔴 Critical Owner: Dev (Dex) Created: 2025-01-14 Updated: 2025-01-17 (v4 - Unified System Integration) Duration: 2.5 days (20 hours) Investment: $250.00.
hatch3r-bug-pipeline
Run a known-cause bug fix through a 3-phase test-first pipeline -- reproduce + root-cause, regression-test + fix together, then root-cause-depth review -- with full sub-agent delegation.
hatch3r-incident-response
Drive a live production incident through a structured lifecycle -- triage + topology, bounded-autonomy mitigation, stakeholder communication, then a blameless post-mortem with runbook -- via delegated sub-agents.