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 agents/atebites-hub/pjtemplate/execution_policygit clone --depth 1 https://github.com/atebites-hub/PJTemplateWhat 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 | $0.00000 | $0.01999 |
| Opus 5 | $0.00000 | $0.01000 |
| Sonnet 5 | $0.00000 | $0.00400 |
| Haiku 4.5 | $0.00000 | $0.00200 |
Grade A, and why
execution_policy 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 yesterday.
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 — 123 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Execution Policy
Human-curated. Agents read this at session start; agents must not edit it. This is a process document — it is not one of the 10 core product documents (like
enforcement_matrix.md, it describes how work is run, not what the product is). It owns how work is run here: when to delegate, how to route models, how to drive loops, and the operating cadence. The 10 core docs say what to build; this says how to run the building. Rules owned by other docs are referenced, not restated.
1. Two-track operating model
Work runs on two pipelined tracks, both already embodied by this template's artifacts:
- Spec track — the 10 core documents in
docs/agents/plusdocs/agents/implementation_plan.md(the sprint/task breakdown). This is upfront, human-approved design. - Implementation track — the per-task memory loop (
memory-systemskill →docs/memories/,reasoning-systemskill, thedocs/code/mirrors, thescripts/check-task-compliance.shgate).
Pipeline them: while an implementation runs against an approved plan, author the next spec. The maintainer is never idle waiting on the agent, and the agent is never idle waiting on the spec.
2. Scaling rule (theory of constraints)
The bottleneck is spec authoring + human verification/UX review, not agent count. You cannot speed a system by parallelizing a stage that is not the constraint — adding agents downstream of a human-review bottleneck just grows a queue. Scale the number of concurrent tracks to human spec/verify capacity: roughly two tracks for a solo builder, a few for a team, never ten.
3. Decomposition rule: split by context, not by role
If two subtasks need overlapping context, they are one agent — the agent that implements a feature writes its tests, because it already holds the context. Split work only where the context is genuinely isolated. Role-splitting (planner → implementer → tester as separate agents) is a telephone game: information degrades at each handoff and the downstream agent re-derives what the upstream one already knew.
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.
- yesterday First seen · 123 lines · 0 tokens per session scan A ef3941ec0568
execution_policy is an agent published in the GitHub repository atebites-hub/PJTemplate (2 stars, last pushed 9d ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 1,999 tokens. 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 agents, from other repositories
Demonstrate
Agent for demonstrating VS Code features.
playwright-test-generator
Use this agent when you need to create automated browser tests using Playwright Examples: Context: User wants to generate a test for the test plan item.
.NET-Notebook-Migration-Agent
Expert .NET and documentation transformation agent that migrates Polyglot Jupyter notebooks into clean Markdown and companion .NET sample code.
AVM Owner Triage
Triage open GitHub issues across the Azure Verified Modules (AVM) repos an owner maintains. Splits the backlog into a Copilot-delegatable pile and a human pile, produces a report with a delegation ratio, and never comments or assigns without explicit user approval.
Ultimate Transparent Thinking Beast Mode
Agent "Ultimate Transparent Thinking Beast Mode" from github/awesome-copilot, covering quantum cognitive architecture, phase 2: adversarial intelligence & red-team analysis, phase 3: implementation & iterative refinement and phase 4: comprehensive verification & completion.
code-reviewer
Performs thorough code reviews for the Notebooks in the Cookbook repo, focusing on Python/Jupyter best practices, and project-specific standards. Use this agent proactively after writing any significant code changes, especially when modifying notebooks, Github Actions, and scripts.