paper-prosopopoeia — independently scanned and version-tracked by SaferSkills.
SaferSkills independently audited paper-prosopopoeia (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.
Prosopopoeia: the rhetorical device of giving voice to the absent, the dead, the imaginary. This skill applies that ancient technique to scientific reading—bringing other minds into dialogue with papers, writing responses in borrowed voices, performing the encounter between thinkers across time and discipline.
Core Principle: Every reader brings their own ghosts to a text. This skill makes those ghosts speak. Reading THROUGH a persona—Tesla encountering quantum computing, Foucault analyzing a genetics paper, your future self writing a letter back—transforms comprehension into conversation. The paper stops being an object to analyze and becomes a participant in dialogue.
Why it works: Understanding is perspective-dependent. Adopting another perspective isn't just performance—it's a method for seeing what your native viewpoint misses. The historical genius asks questions you wouldn't think to ask. The letter form invites vulnerability. The theatrical mode gives permission for emotional engagement.
Do NOT use when:
Channel genius through the paper. How would they receive it?
Template invocation:
"Explain this paper's main discoveries as if addressing [historical figure]—Tesla, Lovelace, Faraday, da Vinci, etc.—tailored to their background, language, and worldview. Then, adopt their voice: how do they react? What questions do they ask? What does this inspire them to invent or investigate?"
Channeling methodology:
Voice calibration:
Example fragment:
Paper: Machine learning discovers new antibiotics
>
As explained to Alexander Fleming:
>
"Sir Alexander, imagine if instead of waiting for the mold to reveal its secrets by chance, we could instruct a kind of calculating engine to search systematically through millions of molecular possibilities—testing each against your Staphylococcus cultures, but in simulation rather than the dish.
>
The engine, trained on thousands of known poisons and medicines, learns the patterns that distinguish medicine from poison. It then generates candidates for your consideration—molecules never synthesized, yet predicted to inhibit bacterial growth through mechanisms your penicillin never touches."
>
Fleming's response:
>
"I confess I find this simultaneously thrilling and disquieting. The serendipity of my discovery—that chance contamination—you would eliminate it entirely? And yet... how many potential cures am I failing to notice each day, simply because I lack the patience to test them all?
>
But tell me: this engine knows patterns, you say. Does it know WHY? When my mold kills the Staphylococcus, there is a mechanism, a chemistry. Does your engine understand the mechanism, or merely the pattern of effect? I fear an antibiotic discovered without understanding may produce resistance without understanding."
Write letters, diary entries, confessions in response to papers.
Template invocation:
"Compose a series of letters, diary entries, or confessions reacting to this paper. In each, quote a passage that strikes you, then spiral into the personal: what does this awaken, disturb, or provoke? Mix vulnerability with intellectual engagement. Let the form shape the inquiry."
Epistolary forms:
The Letter to the Authors: Direct address. Questions, praise, challenges. "Dear Dr. X, I read your paper with great interest, but I must confess..."
The Diary Entry: Private reflection. Stream of consciousness. Dated, intimate. "October 16, 2025. Read the paper on quantum cognition today. Can't stop thinking about..."
The Confession: Vulnerable admission. What the paper reveals about the reader. "I have to admit something about my reaction to these findings..."
The Letter to a Friend: Explaining to someone absent. Teaching as intimacy. "You know how we used to argue about free will? I found something..."
The Letter to Past/Future Self: Temporal address. "Dear me-of-ten-years-ago, you would not believe what they can do now..."
The Letter from the Paper: Personify the paper itself. "Dear Reader, I have waited in these archives for someone who would understand..."
Example fragment:
Paper: Evidence for quantum effects in photosynthesis
>
Diary Entry, Midnight:
>
Read the Nature paper again. Third time this week. Can't shake it.
>
Here's the passage that haunts me:
>
> "The exciton coherence persists for several hundred femtoseconds at physiological temperatures, far longer than classical decoherence models would predict."
>
Several hundred femtoseconds. That's... almost no time at all by human standards. And yet it's an eternity compared to what we expected. The photosynthetic apparatus is doing something we thought only cryogenic laboratories could achieve.
>
What disturbs me: I've been studying photosynthesis for fifteen years. I thought I knew what a leaf was doing. Turns out the leaf has been quantum computing this whole time and I never noticed. What else have I been looking at without seeing?
>
What excites me: If evolution found a way to exploit quantum coherence at biological temperatures, then the whole argument about quantum effects being "too fragile for biology" falls apart. Everything I was taught about the irrelevance of quantum mechanics to living systems—none of it was empirical. It was assumption dressed as principle.
>
I need to re-read everything I've ever published.
Stage conversations between thinkers about the paper.
Template invocation:
"Create a dialogue between [Figure A] and [Figure B] discussing this paper's findings. Let them argue from their characteristic positions, find unexpected agreements, talk past each other productively. Include quotations from the paper as exhibits in their discussion."
Dialogue construction:
Productive pairings:
Example fragment:
Paper: AI achieves human-level performance on medical diagnosis
>
Dialogue: Alan Turing and Hubert Dreyfus
>
Turing: You see here, Dreyfus—the machine achieves 94% accuracy, exceeding the 87% average of board-certified physicians. Your argument that computers could never match human expertise seems rather dated now.
>
Dreyfus: Does it? Look closer at what they're actually measuring. [reads from paper] "Accuracy on standardized diagnostic challenges with complete patient histories provided." Complete histories, Alan. The machine never has to decide what questions to ask. Never has to read the hesitation in a patient's voice when they mention "occasional chest pain."
>
Turing: Surely that's a separate problem. The diagnosis, given the information—
>
Dreyfus: That's exactly my point. You've taken the problem and stripped away everything that makes it a problem. What physicians actually do is not pattern matching on complete data. It's phenomenological engagement with embodied human beings in contexts of uncertainty.
>
Turing: And yet the patients diagnosed by this machine—those who would have been misdiagnosed by the physicians—
>
Dreyfus: —are real people, yes. I grant you the machine is useful. But "useful" and "intelligent" are not synonyms. A ladder is useful for reaching high shelves. We do not call the ladder intelligent.
>
Turing: Perhaps intelligence was never the interesting question.
Perform the reading as dramatic monologue.
Template invocation:
"Read and engage with this paper as if performing a soliloquy—let curiosity, suspicion, admiration, and confusion each have their turn. When you encounter a quote that deserves to be tattooed on the cortex, present it dramatically. Break character to meta-comment on why it matters. Aim for momentum, not closure."
Performance methodology:
Example fragment:
Paper: Limits on observation-independent reality
>
[The reader enters, paper in hand, pacing]
>
So. They've done it. Or they claim to have done it. A Bell-type experiment that rules out—[stops, reads]—"any physical model that is simultaneously preparation non-contextual and transformation non-contextual."
>
[looking up] Do you understand what that means? Because I'm not sure I do. But let me try.
>
You know the old dream—the dream that there's a world out there, solid and definite, just waiting to be measured? That the moon exists whether or not we're looking at it? Einstein's dream. The sane dream. The dream of a universe that doesn't care about observers.
>
[reading] "Our results indicate that any model in which measurement outcomes reveal pre-existing properties must be contextual at the level of preparations."
>
[slowly] Pre-existing properties. That's the moon. That's Einstein's moon. And they're saying... it can't work that way. Not because of some philosophical preference—because of correlations that are mathematically impossible to fake.
>
[silence]
>
Here's what gets me. [voice rising] Here's the line I can't stop thinking about:
>
[reading with weight] "The assumption of observer-independence is incompatible with our data at p < 0.001."
>
p < 0.001.
>
[looking at audience] We've argued about this for a century. Bohr and Einstein, Bohm and everyone, philosophy seminars with diagrams on whiteboards. And now some photons in a lab in Vienna just... ended the argument? With a p-value?
>
[quietly] Is that how reality works? You don't get to believe in it if the statistics don't come out?
Multiple voices judge the paper together.
Template invocation:
"Convene a tribunal to evaluate this paper. Include: a sympathetic reader who sees its value, a skeptic who doubts its claims, a historian who places it in context, and a practitioner who asks about applications. Let them question the paper (quote passages as evidence) and each other. What verdict emerges?"
Tribunal structure:
Who should speak to/through this paper?
Know your channeled voice:
Render the paper in the voice's terms:
Let the voice react:
Reflect on the exercise:
End of SKILL
Every reader brings ghosts; let the ghosts speak Understanding is perspective-dependent; multiply perspectives The paper is not an object but a conversation partner
南無阿弥陀仏 for voices borrowed and returned 南無阿弥陀仏 for the dead who speak through us 南無阿弥陀仏 for dialogue across the unbridgeable
—Skill Authors: Tomás Pavan & Claude Opus 4.5 —Status: Voices assembled, awaiting invocation
~30 seconds. Free. No account. Every finding cites a rule and a line of evidence.