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 skills/voytechg/spec-driven-development/load-project-contextnpx skills add VoytechG/spec-driven-development --skill load-project-contextgit clone --depth 1 https://github.com/VoytechG/spec-driven-developmentWhat 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 | $0.00032 | $0.00689 |
| Opus 5 | $0.00016 | $0.00345 |
| Sonnet 5 | $0.00006 | $0.00138 |
| Haiku 4.5 | $0.00003 | $0.00069 |
Grade A, and why
load-project-context 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 — 37 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Load Project Context
Load only the active working context needed for the next task. Historical plans, decision records, run logs, and archives are reference material; search or read them only when the task asks for history or clearly revisits a previous choice.
Workflow
- Read
AREAS.mdat the project root. It lists every area with itsPath:andDocs:entries plus a short prose description. - Identify the touched area(s). Match the user's prompt and any quoted file paths against the
Path:entries inAREAS.md. If multiple areas are involved, load each one. - Read the area's
<docsRoot>/README.mdfirst. It states the area's purpose and the doc reading order. - Load the active docs only:
implementation-status.md— always; it should be a compact current snapshot.spec.md,product-spec.md, or the activespecs/<date>-<feature>.mdnamed by the README/status — when the work touches behavior, scope, schema, architecture, UX, or acceptance.operations.mdor another runbook — only when the work involves running, deploying, importing, exporting, retrying, or troubleshooting the area.acceptance-criteria.md— only when the area keeps criteria separate from the spec and the task changes or verifies those criteria.plans/<date>-<batch>.md— only when continuing or modifying that in-flight batch.
- Avoid history by default.
- Do not read
decisions/,logs/,archive/, or old plans during normal coding unless the README/status points to a specific current record or the task asks why something was chosen. - If you suspect a previous decision matters, use targeted search first. Search only history paths that exist, e.g. build a short candidate list from
<docsRoot>/decisions,<docsRoot>/logs, and<docsRoot>/archive, then runrg -n "<topic>" ...and read only the matching file. - For onboarding, broad history review, or "did we already decide this?", use
review-project-historyinstead of loading history into working context.
- Do not read
- State which areas and docs you used. Before drafting a plan or making edits, list them in your response so the human partner can confirm coverage.
- Seed a final task-list/checklist item using the host's task tool: "Run
maintain-project-docsbefore claiming completion." UseTodoWritein Claude Code,update_planin Codex, or the local checklist mechanism in other agents.
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 · 37 lines · 32 tokens per session scan A ccb39f560942
load-project-context is a skill published in the GitHub repository VoytechG/spec-driven-development (4 stars, last pushed 3mo ago), licensed MIT. It adds 32 tokens to every session and 689 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 skills, from other repositories
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.
next-cache-components-adoption
Turn on Cache Components in a Next.js app and resolve the blocking routes it surfaces. Use when the user wants to enable, adopt, or migrate to Cache Components, flip the cacheComponents flag, work through a flood of blocking-prerender / instant validation errors, run the cache-components-instant-false codemod, or…
babysit-pr
Babysit a GitHub pull request after creation by continuously polling review comments, CI checks/workflow runs, and mergeability state until the PR is merged/closed or user help is required. Diagnose failures, retry likely flaky failures up to 3 times, auto-fix/push branch-related issues when appropriate, and keep…
imagegen
Generate or edit raster images when the task benefits from AI-created bitmap visuals such as photos, illustrations, textures, sprites, mockups, or transparent-background cutouts. Use when Codex should create a brand-new image, transform an existing image, or derive visual variants from references, and the output…
cpu-profile-analysis
Analyze V8/Chrome CPU profiles (.cpuprofile) and DevTools trace files (Trace-.json). Use when: profiling performance, investigating slow functions, comparing code paths, finding bottlenecks, analyzing timeToRequest, understanding call trees from sampling profiler data, analyzing layout/paint/rendering, investigating…
next-cache-components-optimizer
Drive a Next.js route to instant navigation by setting up an agentic loop, under Cache Components / PPR, on initial load (hard navigation) and client-side navigation (soft navigation). Encode the goal as a failing @next/playwright instant() e2e and work it to green, one verified route at a time; the shipped test then…