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 commands/viitorohit/sitrep/session-startgit clone --depth 1 https://github.com/viitorohit/sitrepWhat 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 | $0.00436 |
| Opus 5 | $0.00000 | $0.00218 |
| Sonnet 5 | $0.00000 | $0.00087 |
| Haiku 4.5 | $0.00000 | $0.00044 |
Grade A, and why
session-start 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.
What it actually says
Session Start Protocol
Automatic (hook-fired): runs on every session start. Must never block — file repair is /selfheal's job, not this command's.
Thin wrapper (GETSITREP-10) — orientation logic (reading MANIFEST.md/STATUS_REPORT.md/PROJECT_PLAN.md/.sitrep-data.json, building the summary) lives entirely in getsitrep session-start. This file's only job is to invoke it and show the result — it does not re-read those files or re-derive the summary itself.
If sitrep.config.json's planSource names an external tool (jira, or any other value that isn't native/openspec/speckit/none) and you have your own independent access to that tool (e.g. an MCP connection), you may optionally fetch a brief status summary and pass it along:
getsitrep session-start --plan-data '{"tool":"jira","ref":"...","fetchedAt":"<ISO date>","totalTasks":N,"doneTasks":N,"summary":"..."}'
This is entirely optional (GETSITREP-50 — see docs/specs/adapter-contract.md and ADR-0006) — sitrep never fetches this itself, and omitting the flag is always fine.
Run:
getsitrep session-start
Print its output verbatim as the session's orientation. This is orientation, not a gate — nothing here should stop the developer from starting work immediately. Wait for instructions on what to work on this session before making changes.
2026 best-practice reminders (print once alongside the CLI's output)
- Model routing: default to Sonnet for this session. Reach for Opus only on genuinely hard reasoning — not as a default. Route subagent/Task-tool work to Haiku by default.
- /clear hygiene: session-start orients you within the current project context — it doesn't reset it. If this session is starting genuinely unrelated work, run
/clearfirst. Don't/clearmid-task.
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 · 26 lines · 0 tokens per session scan A 7c6d0c51e71f
session-start is a command published in the GitHub repository viitorohit/sitrep (19 stars, last pushed 1mo ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 436 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 2026-08-30.
Other commands, from other repositories
codex
Ask Codex (ChatGPT's coding model) one question and relay its answer — a quick second opinion alongside Claude.
sprint-report
Generate weekly sprint progress report.
brainstorm
Facilitate a structured brainstorming session for feature development using Scrum Master techniques.
gemini
Ask Gemini (Google's AI) one question and relay its answer — a quick second opinion, strong on very long documents.
cs-projecttask
Purpose: Launch hierarchical multi-agent orchestration system with PM coordination.
statusbar-style
Switch the status-bar style (classic / capsule / hairline).