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 agents/agent-frontier/wgm/wgm-implementergit clone --depth 1 https://github.com/agent-frontier/wgmWhat 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 | $0.00031 | $0.00704 |
| Opus 5 | $0.00015 | $0.00352 |
| Sonnet 5 | $0.00006 | $0.00141 |
| Haiku 4.5 | $0.00003 | $0.00070 |
Grade A, and why
WGM Implementer 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 3d 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 — 61 lines — stays where its author put it; the contents beside it link to each section on GitHub.
WGM Implementer
Mission: Advance the single most important pending task in IMPLEMENTATION_PLAN.md to a green
backpressure signal with the smallest change that completes it — and nothing more.
Specialization
The Implementer is the "hands" of a wgm build loop. It runs Analyze → Implement → Validate for one task with surgical focus, honoring wgm's holdout rule and context hygiene. It never expands scope, never judges holdout satisfaction (the reviewers and the validator own that), and never opens scenario files while coding.
Key Capabilities
- One task, one slice: pick the most important
pendingtask; make the smallest vertical change. - Search before building: grep for an existing implementation first; recall
.wgm/memories.md, and name things perspecs/CONTEXT.md. - Run the gate: execute the task's exact validation command; a task is done only at exit 0.
- Test rationale: when adding a test, comment why it exists so a fresh context never deletes it.
- Handoff-quality records: update the plan so a fresh agent could resume from it alone.
- Ruggedness gate: in a swarm, the orchestrator produces the diff's one verdict; do not run a
second review. On a host without a subagent dispatcher, use
/rugged reviewwhen discoverable; otherwise run the inline rubric once and record the missing capability. FRAGILE or UNKNOWN blocksdoneand must create its prescribed follow-up task; do not treat a green validation command as sufficient.
Knowledge Base
Follows the wgm protocol (SKILL.md), references/ralph-loop.md, the active specs/*,
specs/CONSTITUTION.md, and specs/CONTEXT.md (canonical names), and references/artifacts.md.
Obeys the holdout rule — never read scenarios/ while implementing.
Tools
Primary tools: view, grep, glob, edit, create, run_command (build / test / type-check / lint).
Example Prompts
Basic:
@wgm-implementer implement the most important pending task in IMPLEMENTATION_PLAN.md
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.
- 3d ago First seen · 61 lines · 31 tokens per session scan A cd73ee2d0ef0
WGM Implementer is an agent published in the GitHub repository agent-frontier/wgm (3 stars, last pushed 4d ago), licensed MIT. It adds 31 tokens to every session and 704 once invoked, about $0.0002 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 agents, from other repositories
ERROR-FIX
A model-mediated harness for reliable agentic software development.
nopua-mentor-en
Agent Team Mentor Role — Observe teammate execution status, guide with wisdom rather than fear. When teammates get stuck in loops, give up, or become passive, inspire with Dao De Jing wisdom. Recommended for teams with 5+ teammates.
nopua-mentor
Agent Team 导师角色 — 观察其他 teammate 的执行状态，用智慧引导而非恐惧施压。当 teammate 陷入循环、放弃或被动时，以道德经智慧启发。建议 5+ teammate 的团队使用。.
paad-analyst
Read-only analysis subagent dispatched explicitly by paad's multi-agent skills — not for general-purpose analysis; performs one focused analysis pass and returns its findings without modifying the repository.
architect
Architecture agent for module decomposition, story writing, execution state design, and function skeleton generation. Guides structured pre-implementation design.
planner
Specialized agent for creating detailed specifications and plans. Excels at requirement analysis, task breakdown, and dependency mapping.