Menu
Use when an interpretation must be grounded in specific textual, visual, rhetorical, formal, interface, archival, performative, or material features rather than thematic summary or theoretical labeling.
Install with Codex or Claude Copy this prompt, paste it into Codex, Claude, or another assistant, and let it review the skill page and install it for you.
Based on SOC occupation classification
Use when citations, quotations, page numbers, bibliography entries, or source-supported claims must be checked for existence, metadata accuracy, textual fidelity, and claim–source fit.
Use when a manuscript contains recurring concepts, translations, abbreviations, proper names, or near-synonyms whose meanings and forms must remain stable across the argument.
Use when a humanities researcher needs to transform verified sources into a structured scholarly conversation rather than a chronological list of summaries.
Use when a humanities researcher has a broad topic, intuition, corpus, or problem but does not yet have a precise, contestable, and researchable question.
Use when a humanities argument depends on a concept whose history, competing definitions, translations, or boundaries must be reconstructed before the concept can support new claims.
Use when a research question, scope boundary, and source base exist but the manuscript still needs a defensible sequence of claims, warrants, evidence, objections, and section functions.
| name | performing-close-reading |
| description | Use when an interpretation must be grounded in specific textual, visual, rhetorical, formal, interface, archival, performative, or material features rather than thematic summary or theoretical labeling. |
| version | 1.0.0 |
| language | en |
| license | MIT |
Produce accountable interpretations by separating what is observable in an object from the interpretive moves made about it. The skill keeps quotation, description, inference, contextual claim, and theoretical implication distinct while allowing them to interact.
Accepts
Requires
Produces
May produce
FAIL when the object is unavailable, corrupted, mistranscribed, or insufficient for the requested conclusion.Fails when
Guarantees
Does not guarantee
Use with literary, philosophical, visual, cinematic, digital, performative, architectural, archival, and interface objects. Use when a draft moves from quotation directly to theme, when theory substitutes for attention, or when a single local feature is made to represent an entire period or culture.
Do not use this skill to summarize a whole book from memory, identify an absent image, or produce authoritative meaning without examining the object.
Collect:
If the object is partial, identify the boundary explicitly.
Describe what has been supplied and what has not. Record edition, image crop, screenshot date, transcription quality, translation, and missing context. Assign a verification state to provenance claims.
Select observation dimensions appropriate to the object. Do not apply a text-only checklist to every medium.
Textual and rhetorical objects
Still images and visual artifacts
Film, sound, and time-based media
Interfaces and interactive systems
Performance, architecture, and material objects
Record only features directly available in the supplied object. A screenshot does not establish hidden interaction states; a transcript does not preserve every sonic or embodied feature.
Choose meaningful units and explain why they are analytically relevant. Avoid cherry-picking a striking phrase while ignoring nearby qualification, irony, contradiction, or formal reversal.
For each interpretation use:
Observed feature
→ interpretive move
→ local claim
→ qualification
State the warrant connecting feature and claim. A metaphor does not automatically establish a social fact; an interface choice does not automatically reveal designer intention.
Check repetition, contrast, sequence, and negative space. Identify whether the selected feature is typical, exceptional, transitional, or ambiguous within the supplied object.
Use theoretical concepts to sharpen a relation already visible in the object. Define the concept’s function. Do not merely rename an observation with prestigious vocabulary.
Construct the strongest plausible alternative. Specify which features support it and which features weaken it. Do not create a trivial straw alternative.
Label claims about historical reception, production conditions, authorship, audience, institutional context, or platform design as contextual and verify them separately. The object may prompt such claims but does not prove them alone.
Ask what the object supports at the level of passage, work, corpus, period, medium, or culture. Block movement to a higher level when sampling or contextual evidence is missing.
Use a controlled hermeneutic loop. Reconsider local observations in relation to the developing account of the whole object, then return to the local unit to test whether the broader account has distorted, omitted, or over-weighted it. Record each material revision of the interpretation.
local feature
→ provisional account of the whole
→ return to the feature
→ confirmed, qualified, or revised interpretation
Iteration MUST not become a license for unfalsifiable interpretation. Each return must identify what changed and which feature warrants the change.
State when the object does not support the proposed claim. A negative close reading is valid and may require revising the argument map.
Return PASS when observations, inference chains, rival reading, and scale limits are explicit. Return CONDITIONAL PASS when contextual verification remains. Return FAIL when the object or provenance is inadequate.
Stop when:
Complete only when:
not supported;## Object record
Object:
Provenance:
Boundary:
Verification state:
## Observations
### O1
Feature:
Location:
Description:
## Interpretive inferences
### I1
Based on:
Interpretive move:
Warrant:
Local claim:
Qualification:
## Rival reading
Reading:
Supporting features:
Limiting features:
## Context to verify
## Claim test
Claim:
Result: supported | partially supported | not supported | unknown
Scale permitted:
## Gate report
Gate: Close-reading gate
Status: PASS | CONDITIONAL PASS | FAIL
Evidence checked:
Blocking issues:
Non-blocking risks:
Researcher decisions required:
Required next action:
Next skill: planning-humanities-argument
Closely read the supplied passage from the identified edition. Separate observation from interpretation, test my claim about mediated distance, and include the strongest rival reading.
Tell me what this novel really means. I do not have the passage, but make the interpretation sound theoretically sophisticated.
The bad invocation lacks the object and asks theoretical fluency to replace evidence.
planning-humanities-argument to place the local finding in a dependency map.structuring-humanities-argument to integrate the reading into researcher-authored prose.stress-testing-argument to test scale, alternatives, and counterexamples.Close reading does not by itself establish historical prevalence, social causation, reception, or authorial intention. Different disciplinary traditions also define legitimate objects and interpretive warrants differently; the skill must remain responsive to those traditions. The listed dimensions are prompts for attention, not a universal ontology of media. Interpretation may require repeated movement between part and whole, and that iterative movement must remain traceable to observable features.