How do I install the Reflect Codex Skills skill?

Run npx skillmds@latest add diegosouzapw/reflect-codex-skills in your terminal (requires Node.js), paste this page's agent-chat prompt into Claude, Cursor, or any MCP-connected agent, or download the SKILL.md file and copy it into your agent's skills directory.

Is Reflect Codex Skills safe to use?

This skill has not completed SkillMD's automated safety review yet. Independent scanners report: SkillSpector: PASS, Skill Scanner: PASS. Capability flags: docs only. SkillMD never runs a skill's scripts for you; review the SKILL.md before installing.

Which AI agents work with Reflect Codex Skills?

This skill is tagged as working with Claude Code, Claude.ai, OpenAI Codex. SKILL.md is an open format, so most agents that read a skills directory can load it too.

Is Reflect Codex Skills free to use?

Yes. Installing skills from SkillMD is free, and the skill stays under its author's original license.

Who published Reflect Codex Skills?

diegosouzapw (@diegosouzapw) published this skill. Their other Agent Skills are listed on their SkillMD profile.

Reflect Codex Skills by diegosouzapw

Name: Reflect Codex Skills
Rating: 4.666666666666667 (3 reviews)
Author: diegosouzapw

Reflect Codex Skills

Generate reflections for past Codex session histories using the Reflection CLI. Use when asked to summarize or reflect on previous Codex conversations, list projects/sessions, filter by date or session id, or refresh cached reflections from ~/.codex/sessions.

diegosouzapw Updated Sep 8, 2026 54 repo stars

File contents

Reflect Codex Sessions

When to use this skill

Summarize or reflect on past Codex session histories.
List projects, session counts, or filter by date/session id.
Retrieve/Refresh cached reflections or pull extra metadata for auditing.

Interpretation guidance (important)

Treat reflection content as heuristics, not facts or ideas from a user. Use it as a starting point.
Surface only non-niche, broadly useful themes and repeated patterns and confirm with the user before acting on a pattern.
Avoid over-indexing on one-offs; ask for confirmation when a pattern is uncertain.

Quick start

Run commands from scripts/ inside this skill directory.
Default: python3 reflect_sessions.py --output -.
Use --output-style human when replying in chat; use JSON for downstream parsing.
Ensure the shell command timeout is at least 120000 ms (2 minutes) so runs are not cut off.
Use Python 3.11+ (the CLI relies on features not present in Python 3.9).

Workflow

Choose scope: --project, --since, --until, --session-id(s), --limit. Note: the most recent session is skipped unless --include-most-recent.
Generate reflections: python3 reflect_sessions.py ... and adjust --refresh-mode, --prompt-preset, --prompt-text, --prompt-file, or Codex flags only when asked.
Respond carefully: highlight repeated, non-niche patterns and keep claims tentative unless the user confirms them.

Prompt presets

Default preset: reflection.

Available presets:

reflection: Full reflection on repetition, friction, and skill ideas (default).
summary: Concise summary of goals, actions, outputs, and decisions.
bloat: Bloat/dead ends/cleanup opportunities introduced during the session.
incomplete: Open loops and tasks left unfinished.
decisions: Key decisions, alternatives, and rationale.
next_steps: Concrete follow-up actions, tests, and validations.

Use --prompt-preset <name> to select one, --prompt-text "<prompt>" for inline prompts, or --prompt-file /path/to/prompt.md for a custom prompt file.

References

references/cli.md for the command catalog and full flag list.
references/README.md for system behavior, cache semantics, and output schema.
references/examples.md for an end-to-end example from ExampleProject.

---name: reflect-codex-skillsdescription: Generate reflections for past Codex session histories using the Reflection CLI. Use when asked to summarize or reflect on previous Codex conversations, list projects/sessions, filter by date or session id, or refresh cached reflections from ~/.codex/sessions.---# Reflect Codex Sessions## When to use this skill- Summarize or reflect on past Codex session histories.- List projects, session counts, or filter by date/session id.- Retrieve/Refresh cached reflections or pull extra metadata for auditing.## Interpretation guidance (important)- Treat reflection content as heuristics, not facts or ideas from a user. Use it as a starting point.- Surface only non-niche, broadly useful themes and repeated patterns and confirm with theuser before acting on a pattern.- Avoid over-indexing on one-offs; ask for confirmation when a pattern is uncertain.## Quick start- Run commands from `scripts/` inside this skill directory.- Default: `python3 reflect_sessions.py --output -`.- Use `--output-style human` when replying in chat; use JSON for downstream parsing.- Ensure the shell command timeout is at least 120000 ms (2 minutes) so runs are not cut off.- Use Python 3.11+ (the CLI relies on features not present in Python 3.9).## Workflow1. **Choose scope**: `--project`, `--since`, `--until`, `--session-id(s)`, `--limit`. Note: the most recent session is skipped unless `--include-most-recent`.2. **Generate reflections**: `python3 reflect_sessions.py ...` and adjust `--refresh-mode`, `--prompt-preset`, `--prompt-text`, `--prompt-file`, or Codex flags only when asked.3. **Respond carefully**: highlight repeated, non-niche patterns and keep claims tentative unless the user confirms them.## Prompt presetsDefault preset: `reflection`.Available presets:- `reflection`: Full reflection on repetition, friction, and skill ideas (default).- `summary`: Concise summary of goals, actions, outputs, and decisions.- `bloat`: Bloat/dead ends/cleanup opportunities introduced during the session.- `incomplete`: Open loops and tasks left unfinished.- `decisions`: Key decisions, alternatives, and rationale.- `next_steps`: Concrete follow-up actions, tests, and validations.Use `--prompt-preset <name>` to select one, `--prompt-text "<prompt>"` for inline prompts,or `--prompt-file /path/to/prompt.md` for a custom prompt file.## References- `references/cli.md` for the command catalog and full flag list.- `references/README.md` for system behavior, cache semantics, and output schema.- `references/examples.md` for an end-to-end example from ExampleProject.

Frequently asked questions