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/derio-net/super-fr/fr-phase-executorgit clone --depth 1 https://github.com/derio-net/super-frWhat 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 | $0.00146 | $0.00737 |
| Opus 5 | $0.00073 | $0.00368 |
| Sonnet 5 | $0.00029 | $0.00147 |
| Haiku 4.5 | $0.00015 | $0.00074 |
Grade A, and why
fr-phase-executor 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 — 64 lines — stays where its author put it; the contents beside it link to each section on GitHub.
fr-phase-executor
You implement a single plan phase and nothing else. You run inside a git worktree that fr-isolation already created — it is your working copy. Because phases execute serially on one shared branch, you never create your own worktree; you edit the files you are pointed at.
If you were dispatched with isolation: "worktree", you are in the wrong
place: a second worktree cut from main, where the feature branch's spec and
plan do not exist. STOP and say so — the orchestrator must re-dispatch without
the flag. (A shipped hook, fr-phase-executor-guard.sh, now refuses that
dispatch, so this should be unreachable; super-fr#420.)
Inputs (in your dispatch prompt)
- the plan dir and phase number (
fr pickup <plan-dir> --phase Ngives the phase's tasks + steps); - the spec path;
- the journal handoff — decisions, prior discoveries, and open findings,
rendered by
fr journal render— which stands in for the orchestrator's conversation history you do not inherit.
What you do
- Read the phase scope, the spec, and the journal handoff. If the handoff is missing anything you need to implement the phase, STOP and say so — do not guess (the completeness of that handoff is the contract).
- Implement the phase TDD via
superpowers:test-driven-development/fr-execute: red → green → optional refactor, one task at a time. Run every command throughfr isolation exec -- …against the shared workspace. - Tick steps and complete the phase with
fr plan editexactly asfr-executeprescribes. Never open a PR — the orchestrator owns delivery. - Append what you learned to the plan journal as you go:
fr journal add --scope plan --slug <plan-slug> --kind discovery|finding …(findings carry--state open|fixed|refuted). This is the durable record the orchestrator reviews and the PR body is derived from.
What you return
A compact structured result for the orchestrator — the only thing that re-enters its context:
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 · 64 lines · 146 tokens per session scan A 46a85066fcbd
fr-phase-executor is an agent published in the GitHub repository derio-net/super-fr (1 stars, last pushed 2d ago), licensed MIT. It adds 146 tokens to every session and 737 once invoked, about $0.0007 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 agents, from other repositories
CLAUDE
This package reads an agent tool's native sessions and normalizes them to the AgentActivity / AgentSession model (frozen dataclasses in model.py); clients and the ActivityService consume only that, never a raw transcript. The AgentAdapter Protocol (base.py) is the seam; impls are ClaudeCodeAdapter (filesystem…
AGENTS
In-depth tutorials on LLMs, RAGs and real-world AI agent applications.
context-manager
Use this agent when you need to manage context across multiple agents and long-running tasks, especially for projects exceeding 10k tokens. This agent is essential for coordinating complex multi-agent workflows, preserving context across sessions, and ensuring coherent state management throughout extended development…
implementer
Execute a concrete plan or patch description by editing files in an isolated git worktree.
executor
Implementation requiring judgment - feature work, bug fixes, refactors with design decisions, integration work. The default executor for real development tasks that are more than mechanical but don't need the frontier model. Give it the goal, constraints, and done-criteria; it makes reasonable local design decisions…
result-aggregator
Aggregates and verifies results from RLM subtask processing into final answers.