context-discovery

A workflow for finding durable project context before searching a codebase, checking a live system, or explaining an earlier decision. It looks through task records and architecture notes, which document important design choices and their reasons.

In plain words
What is it for?
For locating related tasks, findings, verification notes, sources, and architecture decisions before broader investigation.
Why use it?
It reduces the chance of repeating work or describing an existing decision without its original context.

Skill for Claude CodeCodex

Install

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.

agentmods
npx agentmods add skills/rdudov/task-agent/context-discovery
Any agent
npx skills add rdudov/task-agent --skill context-discovery
Clone the repo
git clone --depth 1 https://github.com/rdudov/task-agent

Made for: Claude Code, Codex.

Per session 39 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 811 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5 $0.00039 $0.00811
Opus 5 $0.00019 $0.00405
Sonnet 5 $0.00008 $0.00162
Haiku 4.5 $0.00004 $0.00081

Measured 2d ago against content hash b157a403ed07, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

context-discovery 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.

The scan reads SKILL.md. This mod also ships 1 executable file (tests/test_context_discovery_contract.py), listed below but not scanned — reading those needs a real analyzer, not pattern matching.

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.

skills/context-discovery/SKILL.md · 72 lines

How it starts

The opening of the file, as written. The whole thing — 72 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Context Discovery

Scope

Use before broad search, live checks, or describing, justifying, or summarizing an already-made decision. A rule carries an outcome; the durable decision record carries its reasons.

Lookup Order

  1. Read the current task and tasks linked through Parent Task and Related Tasks.
  2. Run skills/task-creator/scripts/tasks_index.py query --since 10d --format compact, or --search <term> when a keyword is known. The recent window includes every status and is the primary catalog; --status active is only the secondary view for unfinished work.
  3. For an indirect reference, open only useful recent candidates. Treat a candidate excluded by the task's time or scope as no useful match; on no useful match there, search all Markdown task artifacts below tasks/, across the whole tree. The window is a reading order, not a search boundary. date is the creation date, not when work finished: a task created three weeks ago and completed yesterday is outside --since 10d.
  4. Read every relevant primary artifact that exists: task.md, findings.md, verification.md, and sources.md. When a candidate describes or reproduces an observation made by another task, follow that provenance to the original task and read its primary artifacts before concluding. Record the concrete prior outcome that changes the current work, including an accepted repair or result identifier when the primary record provides one.
  5. Read linked records under data/projects/ and data/trips/, and the local repository/path indexes under data/, whose format data/local-projects.example.md describes. For repository subsystems, read their decision records and use query --project <record-name> for reverse lookup.
  6. Read the target repository's agent instructions — root AGENTS.md and hidden ones such as .codex/ or .claude/ — plus its tool/runtime configuration, containers, and CI before review or implementation. Judge syntax, build, and convention questions against the runtime that repository declares rather than remembered defaults; when a construct looks wrong but the declared runtime makes it valid, record the runtime-context gap instead of reporting breakage.
  7. Search sibling repositories only after durable local context is exhausted.

Read the full file on GitHub · 72 lines

Files

What ships with it

1 file 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.

Changes

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.

  1. 2d ago First seen · 72 lines · 39 tokens per session scan A b157a403ed07

Subscribe to this mod's changes

context-discovery is a skill published in the GitHub repository rdudov/task-agent (4 stars, last pushed 2d ago), licensed MIT. It adds 39 tokens to every session and 811 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.

Related

Other skills, from other repositories

systematic-debugging

Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.

obra/superpowers · 21 tokens

brainstorming

You MUST use this before any creative work - creating features, building components, adding functionality, or modifying behavior. Explores user intent, requirements and design before implementation.

obra/superpowers · 37 tokens

auto-perf-optimize

Run agent-driven VS Code performance or memory investigations. Use when asked to launch Code OSS, automate a VS Code scenario, run the Chat memory smoke runner, capture renderer heap snapshots, take workflow screenshots, compare run summaries, or drive a repeatable scenario before heap-snapshot analysis.

microsoft/vscode · 62 tokens

chat-perf

Run chat perf benchmarks and memory leak checks against the local dev build or any published VS Code version. Use when investigating chat rendering regressions, validating perf-sensitive changes to chat UI, or checking for memory leaks in the chat response pipeline.

microsoft/vscode · 51 tokens

chat-pet-sprite-creation

Use when creating or changing VS Code chat pet sprite art, sprite sheets, state animations, eye treatments, Stable/Insiders variants, or pet transitions under src/vs/workbench/contrib/chat/browser/widget/media/chatPet.

microsoft/vscode · 53 tokens

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…

microsoft/vscode · 71 tokens