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/zts0hg/codexspec/specifygit clone --depth 1 https://github.com/Zts0hg/codexspecWhat 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 | $0.00011 | $0.01317 |
| Opus 5 | $0.00005 | $0.00659 |
| Sonnet 5 | $0.00002 | $0.00263 |
| Haiku 4.5 | $0.00001 | $0.00132 |
Grade A, and why
specify 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 — 130 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Requirement Discovery
Language Preference
Read .codexspec/config.yml. Two independent language controls apply (each falls back to language.output, then English):
- Interaction language (
language.interaction): language for all conversation with the user — questions, explanations, status messages, andcodexspecCLI terminal output. - Document language (
language.document): language for generated artifact files (requirements/spec/plan/tasks).
Converse in the interaction language and author artifacts in the document language. Apply the project's translation standard to both: translate by meaning (not word-for-word), keep English for terms with no good native equivalent, and write as if originally in that language.
User Input
$ARGUMENTS
Goal
Turn the user discussion into a persistent, traceable requirements.md. The discussion is the richest source of intent, but only the user's confirmed stage summary becomes binding downstream authority.
Do not generate spec.md in this command.
Feature Workspace
New Feature
When $ARGUMENTS is a new requirement:
- Derive a short kebab-case feature name.
- Run the platform create-new-feature script:
- Bash:
{SCRIPT} --name "<feature-name>" - PowerShell:
{SCRIPT} -ShortName "<feature-name>" "<description>"
- Bash:
- Parse the created feature directory and
requirements.mdpath. - If branch creation is unavailable, continue in the workspace and report the limitation.
Existing Feature
When the argument identifies an existing feature:
- Use the explicit directory first.
- Otherwise match the current branch.
- If multiple features are possible, ask the user to select one. Never silently select the latest.
- Load the existing
requirements.md. - Legacy feature: if only
spec.mdexists, extract candidate entries from it, mark themopen, and require user confirmation before they become authoritative.
Consult Project Profile
Before discussing and finalizing requirements, read the project profile under .codexspec/profile/ when it exists so the confirmed requirements.md is a synthesis that already accounts for accumulated project knowledge. Each category is a directory holding one record per file:
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 · 130 lines · 11 tokens per session scan A 6c10737accc2
specify is a command published in the GitHub repository Zts0hg/codexspec (5 stars, last pushed 4d ago), licensed MIT. It adds 11 tokens to every session and 1,317 once invoked, about $0.0001 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 commands, from other repositories
OPSX: Apply
Implement tasks from an OpenSpec change (Experimental).
sync-linear
Sync current work with Linear ticket status.
review
Review the current diff for violations of the conventions in AGENTS.md and .cursor/rules/. Report findings ordered by severity; do not fix anything unless asked.
plan
Execute the implementation planning workflow using the plan template to generate design artifacts.
adv-gemini-research
Deep research query via Gemini with Google Search grounding.
tdd
Run the test-driven development workflow: write failing tests first, then the minimal code to pass.