Browse reusable Agent Skills, each with a clear purpose and practical guidance.
Use when the sample construction, variable measurement, panel structure, or inference of a Financial Management (FM) manuscript is fragile — before identification can be trusted or exhibits finalized. Hardens the data layer; it does not establish the causal claim (finman-identification) or run robustness (finman-robustness).
Use when the causal or economic credibility of a Financial Management (FM) result is the bottleneck — endogenous corporate policy choices, staggered-event designs, weak instruments, or a "correlation dressed as a channel." Stress-tests the identification to FM's applied-finance bar before exhibits are finalized.
Use when the contribution of a Financial Management (FM) manuscript relative to the published finance frontier is fuzzy, oversold, or indistinguishable from a specialist sibling. Stakes the marginal contribution and the sibling boundary; it does not run analysis or draft the intro prose.
Use when deciding which robustness checks a Financial Management (FM) paper actually needs — FM explicitly puts "less weight on trivial robustness tests," so the craft is selecting threat-targeted checks and demoting busywork. Curates the robustness layer; it does not establish identification (finman-identification) or format exhibits (finman-tables-figures).
Use when deciding whether and how computation appears in a FOCS (IEEE Symposium on Foundations of Computer Science) paper — a venue that accepts on theorems with no evaluation section expected — covering machine-verified case analyses, computer-discovered constructions, and honest illustrative plots.
Use when building the related-work and citation apparatus of a FOCS (IEEE Symposium on Foundations of Computer Science) paper — tracing result lineages across the FOCS/STOC record, citing arXiv/ECCC full versions with version pins, handling concurrent preprints, and keeping self-citation double-blind-safe.
Use when reasoning about how a FOCS (IEEE Symposium on Foundations of Computer Science) submission is evaluated — the TCMF-sponsored program committee, subreferee delegation, double-blind norms the venue adopted years before STOC, the summer decision arc, and what each outcome means for a theory paper.
Use when designing or auditing ESEC/FSE empirical evaluations, covering real subject systems, fair baselines, SE-standard statistics and effect sizes, qualitative and mixed-methods rigor, contamination-aware LLM ablations, provenance for mining studies, and matching evidence to the shape of each software-engineering claim.
Use when reasoning about how an ESEC/FSE research submission is evaluated, covering double-anonymous "heavy" review, the at-least-three-reviewer PC model, the Accept / Major Revision / Reject decision categories, the journal-style revise-and-resubmit round of PACMSE, and how FSE's process differs from ICSE's cycles and ISSTA's rounds.
Use when deciding whether a software-engineering project belongs at ESEC/FSE or should be routed to ICSE, ASE, ISSTA, MSR, ICSME, or an SE journal (EMSE/TSE/TOSEM), and when distinguishing FSE from its sibling general-SE flagship ICSE by contribution shape, evidence maturity, and the PACMSE journal-style fit.