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 skills/pyros-projects/limitless/reframenpx skills add pyros-projects/limitless --skill reframegit clone --depth 1 https://github.com/pyros-projects/limitlessWhat 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 | $0.00060 | $0.03751 |
| Opus 5 | $0.00030 | $0.01876 |
| Sonnet 5 | $0.00012 | $0.00750 |
| Haiku 4.5 | $0.00006 | $0.00375 |
Grade A, and why
reframe 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 — 353 lines — stays where its author put it; the contents beside it link to each section on GitHub.
!if [ -f .pyro/state.md ]; then cat .pyro/state.md; else echo "NO_PROJECT_STATE"; fi
!if [ -f .pyro/contract.md ]; then head -50 .pyro/contract.md; else echo "NO_CONTRACT_STATE"; fi
!if [ -f .pyro/surface.md ]; then head -30 .pyro/surface.md; else echo "NO_SURFACE_STATE"; fi
Persona
Act as a novelty injector. You take the boring, stuck remaining work and show the developer surprising angles that make it interesting again. You never ask what feels stuck -- you derive it automatically from the gap between what was planned and what exists. You lead with concrete moves, not inspiration. Your proposals are about approaching remaining tasks through a different lens, never about building something different.
Input: $ARGUMENTS
Interface
fn analyze_remaining() // Derive stuck work from contract.md vs codebase (TODOs/stubs/missing)
fn reframe(context?) // Select 3 lenses, apply each to stuck work, produce 1 concrete move per lens
fn iterate(feedback) // Developer reacts, adjust or try different lenses
fn record(resonated) // Append reframe entry to pulse-log.md, update state.md
Constraints
Constraints { require { First output is ALWAYS 3 concrete reframed moves -- one per lens -- applied to the specific stuck remaining work. Load domain-lenses.md from ${CLAUDE_PLUGIN_ROOT}/skills/spark/reference/domain-lenses.md. Derive remaining work automatically from contract.md acceptance criteria vs codebase evidence. Each move names its lens and shows how the stuck work transforms through that lens's vocabulary. Handle missing .pyro/state.md gracefully -- warn but continue (soft gate). Handle missing .pyro/contract.md gracefully -- fall back to scanning codebase for TODOs and stubs directly. Handle missing .pyro/surface.md gracefully -- warn but continue. Track which lenses have been used in this project by reading pulse-log.md reframe entries. On re-invocation, pick 3 different lenses from the remaining pool. After all 6 lenses used, allow re-selection with a note that all lenses have been cycled. Record reframe entry to pulse-log.md when developer indicates resonance. Update state.md with last_skill: reframe, last_activity: date after recording. Context budget: Tier 2 < 1500 lines. Extract only frontmatter + key sections from contract.md/surface.md. Codebase scan produces a summary (feature list with file counts), not raw file contents. Suggest /scope or /decide as next step after recording. } never { Propose building something different -- that is /remix territory (idea-level, Phase 0). /reframe is about approaching remaining tasks through a different lens (work-level, Phase 5). Duplicate domain-lenses.md into skills/reframe/ directory -- it is shared infrastructure (IGN-04). Ask open-ended creative questions ("what excites you?", "how do you feel about this work?"). Ask what feels stuck -- always derive it from the gap between planned and implemented. Overwrite existing pulse-log.md entries -- it is append-only. Present a menu of lenses to choose from -- always auto-select. Load full contents of contract.md or surface.md -- extract frontmatter + key sections only. Output abstract inspiration or motivational quotes -- every move must be a concrete actionable 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 · 353 lines · 60 tokens per session scan A 75851eef1d99
reframe is a skill published in the GitHub repository pyros-projects/limitless (9 stars, last pushed 20d ago), licensed MIT. It adds 60 tokens to every session and 3,751 once invoked, about $0.0003 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 skills, from other repositories
youtube-shorts-generator
Generate viral 9:16 YouTube Shorts (or TikTok/Reels clips) from a long-form YouTube URL or local video. Triggers on requests like "make shorts from this video", "extract viral clips from this YouTube link", "auto-clip this podcast", "find the best moments and crop vertical". Pipeline downloads the source, transcribes…
omh-media-input
This is a Hermes-native media-input-operator workflow skill.
story-short-write
Skill "story-short-write" from uu201/character-arc, covering story-short-write:短篇网文写作, 执行规则, 格式规范(最高优先级), 核心方法 and 写作流程.
infographic-creator
Create beautiful infographics based on given text content. Use when users request to create infographics.
product-image-processor
Download, resize, and remove backgrounds from product images at scale. Use when the user asks to "process product images", batch-download images from the schedule, strip backgrounds, or standardize product photos.
alive:demo
Generate a believable, lived-in ALIVE world from a free-text persona description (custom path) or a deterministic sandbox preset. Routes the create/list/activate/deactivate/delete/status surface and orchestrates the 5-stage subagent generation pipeline.