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 instructions/tale-project/tale/agents-mdgit clone --depth 1 https://github.com/tale-project/taleWhat 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 | $0.01501 | $0.01501 |
| Opus 5 | $0.00750 | $0.00750 |
| Sonnet 5 | $0.00300 | $0.00300 |
| Haiku 4.5 | $0.00150 | $0.00150 |
Grade A, and why
tale AGENTS.md 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 — 100 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Working in this repository
The shared contract for every coding agent in every tale-project repository. Read it, then read
.agents/repo.md — the repo-specific contract — before your first change.
Repositories are Bun workspaces (with Turborepo where there is more than one workspace); a
workspace script runs through bun run --filter <workspace> <script>.
Read the repo contract next
.agents/repo.md is the per-repo file: layout, stack, domain rules, extra
gates, and the debt ledger. This shared file never carries repo facts — if a fact is specific to
one repository, it belongs there. When the two seem to disagree, the repo contract wins.
How to work
The biggest quality lever is deciding well, not typing fast. Work in this order:
- Classify the task and follow its discipline end-to-end: a defect (find the root cause, lock it with a regression test); structure-not-behaviour (refactor without changing behaviour); new behaviour (a feature, screen, endpoint, or flag); a review (an adversarial read before merge). Exploring is read-only and returns the conclusion; a migration is phased and reversible, each phase green.
- Write a short planning note first — capture intent, status quo, and plan before any edit; keep scratch files in your global notes directory, never in the clone.
- Unknowns outside the repo? Research before deciding — questions first, sources in order, evidence in the note.
- Search before you write — orient, find the concept to reuse, enumerate the blast radius. The request names one site; the task is the concept.
- UI in scope? Learn the repo's design system first — where it lives is in
.agents/repo.md— then build to it. - Too big for one thread? Split it into disjoint units with complete briefs; you keep the done-gate.
- Do the work thin and reversible, following the discipline you classified — ask the moment a fork or roadblock appears; never guess.
- Prove it — tests carry the change; observe the real outcome; drive web UIs in a real browser.
- Review your own diff — adversarial read, then the automated reviewers.
- Land it — meet the shared definition of done, atomic commits, one focused PR.
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 · 100 lines · 1,501 tokens per session scan A 0a2c1da31f67
tale AGENTS.md is an instructions file published in the GitHub repository tale-project/tale (23 stars, last pushed 2d ago), licensed MIT. It adds 1,501 tokens to every session, about $0.0075 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 2026-08-30.
Other instructions, from other repositories
jcode-ragmir AGENTS.md
AGENTS.md instructions for jcode-works/jcode-ragmir, covering ragmir, working rules, product boundary, privacy and ingestion and documentation.
jcode-ragmir CLAUDE.md
Claude Code instructions for jcode-works/jcode-ragmir, covering ragmir development notes, commands, workspace, boundaries and gitnexus: code intelligence.
nacre CLAUDE.md
Claude Code instructions for nacre-work/nacre, covering nacre — core, the six invariants, never, in any change, commands and layout.
kiln AGENTS.md
AGENTS.md instructions for sequelcore/kiln, covering agents, project, adopted project context, project context and project.
kiln CLAUDE.md
Claude Code instructions for sequelcore/kiln, covering claude project guidance, project, adopted project context, project context and project.
walkerOS CLAUDE.md
Instructions for elbwalker/walkerOS, covering claude.md and important: use agent.md.