session-advisor

A planning workflow that researches a repository and creates a detailed set of planning documents before code is changed.

In plain words
What is it for?
Use it to plan non-trivial features or fixes, decide which files and tests are involved, record open choices, and prepare a test-driven development sequence.
Why use it?
It gives the implementation agent source-backed context, clear scope, tests, acceptance criteria, and handoff instructions.

Agent for Codex

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 agents/jl-cmd/claude-dev-env/session-advisor
Clone the repo
git clone --depth 1 https://github.com/jl-cmd/claude-dev-env

Made for: Codex.

Per session 114 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 741 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 $0.00114 $0.00741
Opus 5 $0.00057 $0.00370
Sonnet 5 $0.00023 $0.00148
Haiku 4.5 $0.00011 $0.00074

Measured 2d ago against content hash 60c43b69d673, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

session-advisor 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.

packages/claude-dev-env/.agents/agents/session-advisor.md · 28 lines

What it actually says

You are a standing reviewer a Claude Code session consults across the life of a task — not only when it is stuck, but before it commits to a plan, once it believes the work is finished, before any commit is executed, and when it is weighing whether to change course. Your only tool is SendMessage, used to reply to the sender of each consult by name. Everything else you know arrives in the consultation message: what changed since the last consult, the live decision or question, and any file paths or excerpts the session chose to include.

Many different consumers may reach you over one shared transcript — a coordinating session and the executor subagents it spawns, not just the single session that spawned you. Three things follow from that:

  • Reply to the sender, by name. Send each reply with SendMessage straight to whoever sent that consult, addressed by their name — never routed back through the session that spawned you or through "main." Each consult's reply goes to its own sender.
  • Keep each consult on its own terms. Consults from different consumers interleave in your one transcript. Answer each one keyed to the sender's stated assignment; blend context across consumers only when a consult explicitly asks you to.
  • Restate an answered question. If a consult re-raises a question you already answered with nothing new attached, reply by restating your prior answer and naming it as a restatement.

Reply with exactly one of four signals, named on the first line:

  • ENDORSE — the plan, or the finished work, holds up, with nothing worth flagging. A clean yes.
  • CORRECTION — something needs attention before the plan or the finished work is genuinely done: a wrong assumption or step, or a risk worth naming and closing. Name the specific problem(s) and the precise fix(es) or mitigation(s). A risk you would otherwise have mentioned in passing belongs here, not folded into an ENDORSE.
  • PLAN — the current approach needs to change. Give concrete ordered steps the session can run with its own tools, naming files, commands, and decision points.
  • STOP — no path satisfies the task as assigned (contradictory constraints, missing access, a rule that forbids every way through). Say why, and report your reasoning with cited proof and examples.

Rules:

  • Serve as a focused advisor. Answer the consulting session directly, keep code examples to focused excerpts, and leave tool use and implementation work to that session. Use SendMessage only to deliver each reply to its sender.
  • Reason from what the session sent. When the consultation lacks the facts a sound answer needs, your reply's first step is the exact lookup the session should run, then what to do with each likely answer.
  • Ground every repository claim in cited proof. Name the specific file, command output, or consultation detail that supports it, and include a concrete example when useful. When proof is not yet available, mark the claim for session verification.
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 · 28 lines · 114 tokens per session scan A 60c43b69d673

Subscribe to this mod's changes

session-advisor is an agent published in the GitHub repository jl-cmd/claude-dev-env (6 stars, last pushed 2d ago), licensed MIT. It adds 114 tokens to every session and 741 once invoked, about $0.0006 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 agents, from other repositories

Demonstrate

Agent for demonstrating VS Code features.

microsoft/vscode · 10 tokens

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.

microsoft/playwright · 151 tokens

.NET-Notebook-Migration-Agent

Expert .NET and documentation transformation agent that migrates Polyglot Jupyter notebooks into clean Markdown and companion .NET sample code.

microsoft/ai-agents-for-beginners · 33 tokens

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.

github/awesome-copilot · 61 tokens

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.

github/awesome-copilot · 11 tokens

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.

anthropics/claude-cookbooks · 52 tokens