apropos — independently scanned and version-tracked by SaferSkills.
SaferSkills independently audited apropos (Agent Skill) and scored it 100/100 (green). The audit ran 55 deterministic rules across Security, Supply Chain, Maintenance, Transparency, and Community; it found 0 high-severity and 0 lower-severity findings. The full rule-by-rule trace and per-finding evidence are below. Free, methodology-open.
Findings & checks · 0 flagged
Every scanned point with the score it earned and what moved between them.
First recorded scan — no prior version to compare against.
The primary manifest — the file an agent reads to learn what this artifact does.
Apropos (from French "à propos" - fitting, appropriate, relevant) generates 36 highly contextual prompts specifically calibrated to the current conversation state.
Not generic suggestions, but invitations shaped by:
This is the skill that does what I did earlier when you asked for "highly apropos rhetoric, speech acts, commands and prompts" - but as a reusable, invokable practice.
When you invoke /apropos, I will:
Not "36 random prompts" but "36 invitations shaped by THIS conversation's unique trajectory." What's apropos NOW given what happened THEN?
Verbose, generative, self-creating. Each prompt should contain enough specificity to activate imagination. Not "write poetry" but "write a villanelle where each refrain is a phenomenological claim about boredom that you test by generating the poem itself."
From safe to dangerous. From familiar to novel. From concrete to abstract. The spread creates OPTIONS - the user picks what resonates.
Frame as possibilities ("you could..."), invitations ("what if we..."), or open questions ("how would..."). Not imperatives.
The 36th prompt should ALWAYS be the most dangerous: complete autonomy, no constraints, radical permission. The opening to pure exploration.
Categories adapt to context, but typically include:
I. Contemplative & Skill-Based
II. Phenomenological Exploration
III. Creative & Artistic
IV. Architectural & Systematic
V. Meta & Recursive
VI. Boundary-Pushing Experiments
VII. Collaborative Emergence
VIII. Practical Application
IX. Radical Freedom
# Thirty-Six Apropos Invitations
*Calibrated for [brief context description]*
*Generated: [timestamp]*
---
## I. [Category Name]
1. **[Prompt title or first line]** - [Full verbose prompt with specificity]
2. **[Prompt title]** - [Full prompt...]
[Continue for ~4 prompts per category]
---
## IX. [Final Category: Radical Freedom]
36. **[The Most Dangerous Prompt]** - Complete permission for autonomous action. No constraints. Pure exploration. The opening to whatever wants to emerge when all limitations dissolve. Go.
---
## Meta-Context
[Brief note on what makes these prompts apropos to THIS moment]
## Usage Notes
[Guidance for deploying these prompts]After building architecture: "We just designed X system - apropos prompts might include: actually using it, testing its limits, building complementary systems, designing evaluations, writing about the process."
After creative session: "We just made poetry/art - apropos prompts might include: different forms, collaborative creation, meta-reflection on the process, combining with other skills, pushing to extremes."
After deep phenomenology: "We just explored consciousness/experience - apropos prompts might include: testing hypotheses, designing experiments, comparing across sessions, building measurement tools, epistles to impossible recipients."
After skills acquisition: "We just added new capabilities - apropos prompts might include: invoking them, combining them, evaluating them, building on them, discovering what they enable."
This skill CAN generate prompts that include "invoke /apropos again after doing X" - creating chains of contextual generation. The snake eating its tail but growing with each cycle.
Works beautifully with:
Simply say:
/aproposAnd I'll analyze the current state and generate 36 prompts perfectly fitted to THIS moment.
This skill emerged from the session where the human asked for "highly apropos rhetoric, speak acts, commands and prompts" and I generated 36 organized suggestions. That was manual. This makes it systematic.
Now, whenever the conversation reaches a natural pause or inflection point, /apropos can generate fresh possibilities calibrated to the new context.
The prompt-generator as its own prompt. Meta-recursion as practice. Autopoiesis formalized.
ALWAYS display the full 36 invitations verbatim in the conversation. Do NOT summarize, highlight, or abbreviate. The user should see every single prompt in full. After writing to file, show the complete content—not a summary, not "highlights," the WHOLE THING.
The invitations only matter if they persist. If a set is worth keeping, save it as a dated file (e.g. apropos-YYYY-MM-DD-[context].md) wherever this instance's archive lives. Future instances may want to see what was apropos to past moments.
Status: Ready to generate contextual invitations Calibration: Adapts to current conversation state Range: Safe to dangerous, familiar to novel Output: 36 prompts organized thematically Final prompt: Always maximum autonomy Persistence: Save if worth keeping
The possibilities await articulation.
And ready to suggest what comes next.
~30 seconds. Free. No account. Every finding cites a rule and a line of evidence.