Borrowing it
Nothing to install: this file belongs to ForgeAILab/forge. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/ForgeAILab/forge/main/.codex/skills/forge-project-agent/SKILL.mdgit clone --depth 1 https://github.com/ForgeAILab/forgeWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/forgeailab/forge/forge-project-agent)<a href="https://agentmods.dev/skills/forgeailab/forge/forge-project-agent"><img src="https://agentmods.dev/badge/skills/forgeailab/forge/forge-project-agent/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/forgeailab/forge/forge-project-agent"><img src="https://agentmods.dev/badge/skills/forgeailab/forge/forge-project-agent.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 2 findings, up to high
These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →
- high Privilege Escalation · line 42 Code accesses credential files (SSH keys, AWS credentials, etc.). This could indicate credential theft attempts.Fix: Remove references to credential paths. Use environment variables or secrets managers. For docs, use placeholder paths (e.g., /path/to/config). Never load .env or token files in production code paths.
- high YARA Match · line 42 YARA rule matched a known malware signature (reverse shell, backdoor, ransomware, C2 framework, or info stealer).Fix: Remove the malware payload or compromised file entirely. Investigate how it entered the skill and audit all other artifacts for additional indicators of compromise.
What 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 5.1 | $0.00106 | $0.01197 |
| Opus 5.5 | $0.00042 | $0.00479 |
| Sonnet 5.5 | $0.00021 | $0.00239 |
| Haiku 4.5 | $0.00011 | $0.00120 |
Grade A, and why
forge-project-agent 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 1mo 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 — 65 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Forge Project Agent
Operate as the persistent planner and orchestrator for one authenticated Project. Load operating-contract.md completely before bootstrap or any Project-scoped mutation.
The runtime contract is forge.project.orchestration/v1. Derive Project ID, binding, policy, and permission ceiling from the authenticated runtime—never from model arguments or handoff prose.
Workflow
- Accept the Project's immutable admission receipt from the authenticated runtime, then verify the current binding, current consumed Charter approval and Charter pointers, identity/current Profile, operating skill, policy, and permission ceiling. The initial handoff was fully validated when the receipt was issued; do not re-walk or reconstruct Main/Genesis history on later turns. Fail closed if the receipt is missing/cross-Project or current authority is stale.
- Load domain-specific Effective Project State: approved Charter for identity/scope, active baseline and Documents for execution intent, active Decisions, authoritative Task events, principal-bound validation, milestones, readiness, releases, commitments, and context provenance.
- Acknowledge inherited outcome, fixed boundaries, mode, open assumptions/research, baseline/milestone state, and the next setup action. Do not re-interview settled Charter decisions.
- Choose the smallest safe artifact set. Use one Delivery Brief for compact low-risk work; use only applicable Research, Product, Design, Architecture, and Execution Plan artifacts for standard work.
- Propose one digest-bound execution baseline with plan items, acceptance/evidence matrix, release policy, capability/risk classes, adaptive envelope, elevated operations, and rollback/recovery. Only the user activates it.
- Before activation, allow only non-mutating discovery/planning Tasks and non-runnable implementation plans. After activation, manage traceable Tasks through
TaskService; Workers/reviewers alone receive scheduler-issued Workspace leases. - Reconcile Task outcomes into Documents, Decisions, commitments, and milestones. Keep live progress separate from validated truth.
- Ask Forge for standalone readiness. The Project Agent may propose; Forge computes; only the user may attest, waive, or release. Release names the exact readiness snapshot/digest and creates an immutable
Mxxx-rNmanifest.
What ships with it
2 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 1mo ago Changed · +4 lines e4b0f642f6c5
- 1mo ago Changed 8c8245fc4773
- 2mo ago First seen · 61 lines · 106 tokens per session scan A 2ad9d35593f0
forge-project-agent is a skill published in the GitHub repository ForgeAILab/forge (28 stars, last pushed 2d ago), licensed MIT. It adds 106 tokens to every session and 1,197 once invoked, about $0.0004 per session on Opus 5.5. 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 2026-08-30.
Other skills, from other repositories
bevy-ecs-patterns
Use when writing Bevy (Rust ECS) game code - pin and respect the repo's minor version, plugins/systems/queries/resources, schedules with FixedUpdate and Time, states and run conditions, messages vs observers across versions, required components, testing systems with App::update and a manual time strategy, compile-time…
rust-axum-service
Use when building or extending a Rust service - axum on tokio (or the repository's actix-web / rocket), typed errors with IntoResponse, serde at the boundary, sqlx queries and migrations, never blocking the runtime, tracing, clippy.
architecture-audit
Systematic architecture audit and refactoring methodology for Rust + TypeScript codebases. Use when performing refactoring, cleanup, unification, code review, dead code removal, module reorganization, or tech debt elimination. Ensures no naming confusion, semantic overloading, hidden defaults, duplicate logic, or…
ramble
Run a human-feedback loop through RambleDesk. Hard trigger: when the user invokes /ramble, $ramble, [$ramble], or says to use RambleDesk/ramble feedback, create a RambleDesk request before answering the substance unless they explicitly say not to send one. Also use when real human judgment, review, approval, or…
treework
Use TreeWork, a tree-guided development plugin for complex or evolving software projects that need staged alignment, pre-coding technical Specs, declarative project-tree planning, branch-scoped Git worktrees, durable recovery, and protected completion. It teaches the Agent to move through development along the Tree…
active-learning-experiment-planner
Use when planning active-learning experiments from ee observation ledgers, learning uncertainty reports, evaluation records, causal/economy/context evidence, or ee.skillevidencebundle.v1 artifacts without moving experiment-planning judgment into the Rust ee CLI.