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/srnichols/plan-forge/context-fuelgit clone --depth 1 https://github.com/srnichols/plan-forgeWhat 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 | $0.00767 | $0.00767 |
| Opus 5 | $0.00383 | $0.00383 |
| Sonnet 5 | $0.00153 | $0.00153 |
| Haiku 4.5 | $0.00077 | $0.00077 |
Grade A, and why
plan-forge context-fuel.instructions.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 — 60 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Context Fuel
Context is what fuels good agent output. Running low starves quality.
When to Load What
| Situation | Action |
|---|---|
| Session start | Call forge_capabilities once — gets the full surface (tools, skills, config, extensions) in one call instead of reading multiple files |
| Working on database code | Prioritize database.instructions.md, security.instructions.md — read these first, reference others only when needed |
| Working on API endpoints | Prioritize api-patterns.instructions.md, auth.instructions.md, errorhandling.instructions.md |
| Working on tests | Prioritize testing.instructions.md — the test patterns and Temper Guards are more valuable than reviewing architecture rules |
| Executing a plan slice | Read the plan's Scope Contract and the specific slice — don't reload the full plan every slice |
| Context feels degraded | Stop and re-read architecture-principles.instructions.md — it's the universal baseline that anchors everything else |
Recognizing Context Degradation
You are losing context when:
- You repeat a mistake that a loaded instruction file already warns against
- You forget a constraint you acknowledged earlier in the session (naming convention, forbidden pattern, tech commitment)
- You make an architectural decision that contradicts the project's established patterns
- You suggest adding a dependency that the Project Principles explicitly forbid
- Your output quality drops noticeably — shorter explanations, missed edge cases, skipped validation
- You can't recall the Scope Contract for the current slice
What to Do When Context Degrades
- Re-read the active instruction files — not all of them, just the ones relevant to the current task
- Check Project Principles — re-read
docs/plans/PROJECT-PRINCIPLES.mdif it exists - Search memory — if OpenBrain is configured, run
search_thoughtsfor the current task context to recall prior decisions - Recommend a fresh session — Plan Forge uses 4 sessions for a reason. If context is severely degraded, suggest the user start a new session and hand off via the plan's status artifacts
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 · 60 lines · 767 tokens per session scan A 9d8a0b910c48
plan-forge context-fuel.instructions.md is an instructions file published in the GitHub repository srnichols/plan-forge (5 stars, last pushed 22d ago), licensed MIT. It adds 767 tokens to every session, about $0.0038 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 instructions, from other repositories
dotnet-skills AGENTS.md
Instructions for managedcode/dotnet-skills, covering agents.md, purpose, solution topology, rule precedence and path and linking rules.
dotnet-skills copilot-instructions.md
Instructions for managedcode/dotnet-skills: Use AGENTS.md as the repository-wide source of truth for workflow, catalog structure, release policy, and skill maintenance rules.
copilot-instructions copilot-instructions.md
Instructions for SebastienDegodez/copilot-instructions, covering copilot instructions, language policy, development code generation and workflow implementation.
Perigon.CLI copilot-instructions.md
Instructions for AterDev/Perigon.CLI, covering github copilot instructions, general guidelines, 技术栈, 项目结构与分层 and 代码风格约定.
maf-doctor maf-deployment.instructions.md
Always-loaded production-deployment patterns for MAF 1.3.0. Auto-applies to Program.cs, DI registration files, and infra config. Covers ManagedIdentityCredential, MaxTokens caps, secret handling, OpenTelemetry wiring, and the analyzer rules that catch regressions at write time.
maf-doctor copilot-instructions.md
Instructions for joslat/maf-doctor, covering maf 1.3.0 migration — auto-loaded constraints, maf 1.3.0 — constraints & breaking changes reference, hard constraints (never violate), fan-out / fan-in rules (silent failure risk) and key breaking changes.